Datasets:

librarian-bots

/

arxiv-metadata-snapshot

like 3

Follow

Librarian Bots 63

2306.06845

Hai-Xiao Wang

Hai-Xiao Wang

Information-Theoretic Limits and Strong Consistency on Binary Non-uniform Hypergraph Stochastic Block Models

The paper is written. New results concern the information-theoretic limits are added. Provide a new refinement algorithm to achieve strong consistency. Thus the title of the paper is changed

math.ST math.PR stat.ME stat.ML stat.TH

http://creativecommons.org/licenses/by/4.0/

Consider the unsupervised classification problem in random hypergraphs under the non-uniform Hypergraph Stochastic Block Model (HSBM) with two equal-sized communities, where each edge appears independently with some probability depending only on the labels of its vertices. In this paper, the information-theoretic limits on the clustering accuracy and the strong consistency threshold are established, expressed in terms of the generalized Hellinger distance. Below the threshold, it is impossible to assign all vertices to their own communities, and the lower bound of the expected mismatch ratio is derived. On the other hand, the problem space is (sometimes) divided into two disjoint subspaces when above the threshold. When only the contracted adjacency matrix is given, with high probability, one-stage spectral algorithms succeed in assigning every vertex correctly in the subspace far away from the threshold but fail in the other one. Two subsequent refinement algorithms are proposed to improve the clustering accuracy, which attain the lowest possible mismatch ratio, previously derived from the information-theoretical perspective. The failure of spectral algorithms in the second subspace arises from the loss of information induced by tensor contraction. The origin of this loss and possible solutions to minimize the impact are presented. Moreover, different from uniform hypergraphs, strong consistency is achievable by aggregating information from all uniform layers, even if it is impossible when each layer is considered alone.

2306.07829

Victor Roca i Lucio

Victor Roca i Lucio

Higher Lie theory in positive characteristic

52 pages. Major expansion of the previous paper, many new results added. Previous results are still included. Corrected minor mistakes, including the formula in Definition 2.16. Comments are very welcome!

math.AT

http://creativecommons.org/licenses/by/4.0/

The main goal of this article is to develop integration theory for absolute partition $L_\infty$-algebras, which are point-set models for the (spectral) partition Lie algebras of Brantner-Mathew where infinite sums of operations are well-defined by definition. We construct a Quillen adjunction between absolute partition $L_\infty$-algebras and simplicial sets, and show that the right adjoint is a well-behaved integration functor. Points in this simplicial set are given by solutions to a Maurer-Cartan equation, and we give explicit formulas for gauge equivalences between them. We construct the analogue of the Baker-Campbell-Hausdorff formula in this setting and show it produces an isomorphic group to the classical one over a characteristic zero field. We apply these constructions to show that absolute partition $L_\infty$-algebras encode the $p$-adic homotopy types of pointed connected finite nilpotent spaces, up to certain equivalences which we describe by explicit formulas. In particular, these formulas also allow us to give a combinatorial description of the homotopy groups of the $p$-completed spheres as solutions to a certain equation in a given degree, up to an equivalence relation imposed by elements one degree above. Finally, we construct absolute partition $L_\infty$ models for $p$-adic mapping spaces, which combined with the description of the homotopy groups gives an algebraic description of the homotopy type of these $p$-adic mapping spaces parallel to the unstable Adams spectral sequence.

2306.07977

Santanu Acharjee

Ahmad Al-Omari, Murad Ozcog, Santanu Acharjee

Primal-Proximity Spaces

Submitted to journal on 23.02.2023

Mathematica, 2024, 66(89), 2, 151-167

10.24193/mathcluj.2024.2.01

math.GN math.FA

http://creativecommons.org/licenses/by/4.0/

The main purpose of this paper is to introduce and study the primal-proximity spaces. Also, we define two new operators via primal proximity spaces and investigate some of their fundamental properties. In addition, we obtain a new topology, which is weaker than old one, via these new operators. Moreover, we not only discuss some of their properties but also enrich with some examples.

2306.08312

Xiang Yu

Lijun Bo, Yijie Huang, Xiang Yu

A decomposition-homogenization method for Robin boundary problems on the nonnegative orthant

Final version, forthcoming in Electronic Journal of Probability

math.PR math.AP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper studies the existence and uniqueness of a classical solution to a type of Robin boundary problems on the nonnegative orthant. We propose a new decomposition-homogenization method for the Robin boundary problem based on probabilistic representations, which leads to two auxiliary Robin boundary problems admitting some simplified probabilistic representations. The auxiliary probabilistic representations allow us to establish the existence of a unique classical solution to the original Robin boundary problem using some stochastic flow analysis.

2306.08918

Runmin Cong

Runmin Cong, Wenyu Yang, Wei Zhang, Chongyi Li, Chun-Le Guo, Qingming Huang, and Sam Kwong

PUGAN: Physical Model-Guided Underwater Image Enhancement Using GAN with Dual-Discriminators

8 pages, 4 figures, Accepted by IEEE Transactions on Image Processing 2023

eess.IV cs.CV

http://creativecommons.org/licenses/by-nc-sa/4.0/

Due to the light absorption and scattering induced by the water medium, underwater images usually suffer from some degradation problems, such as low contrast, color distortion, and blurring details, which aggravate the difficulty of downstream underwater understanding tasks. Therefore, how to obtain clear and visually pleasant images has become a common concern of people, and the task of underwater image enhancement (UIE) has also emerged as the times require. Among existing UIE methods, Generative Adversarial Networks (GANs) based methods perform well in visual aesthetics, while the physical model-based methods have better scene adaptability. Inheriting the advantages of the above two types of models, we propose a physical model-guided GAN model for UIE in this paper, referred to as PUGAN. The entire network is under the GAN architecture. On the one hand, we design a Parameters Estimation subnetwork (Par-subnet) to learn the parameters for physical model inversion, and use the generated color enhancement image as auxiliary information for the Two-Stream Interaction Enhancement sub-network (TSIE-subnet). Meanwhile, we design a Degradation Quantization (DQ) module in TSIE-subnet to quantize scene degradation, thereby achieving reinforcing enhancement of key regions. On the other hand, we design the Dual-Discriminators for the style-content adversarial constraint, promoting the authenticity and visual aesthetics of the results. Extensive experiments on three benchmark datasets demonstrate that our PUGAN outperforms state-of-the-art methods in both qualitative and quantitative metrics.

2306.13977

Li Chen

Zhao Li, Jing Zhang, Jiqiang Zhang, Guozhong Zheng, Weiran Cai and Li Chen

A Generative Hypergraph Model for Double Heterogeneity

11 pages, 6 figures

Journal of Complex Networks 12 (1), cnad048 (2024)

10.1093/comnet/cnad048

physics.soc-ph cond-mat.dis-nn nlin.AO

http://creativecommons.org/licenses/by/4.0/

While network science has become an indispensable tool for studying complex systems, the conventional use of pairwise links often shows limitations in describing high-order interactions properly. Hypergraphs, where each edge can connect more than two nodes, have thus become a new paradigm in network science. Yet, we are still in lack of models linking network growth and hyperedge expansion, both of which are commonly observable in the real world. Here, we propose a generative hypergraph model by employing the preferential attachment mechanism in both nodes and hyperedge formation. The model can produce bi-heterogeneity, exhibiting scale-free distributions in both hyperdegree and hyperedge size. We provide a mean-field treatment that gives the expression of the two scaling exponents, which agree with the numerical simulations. Our model may help to understand the networked systems showing both types of heterogeneity and facilitate the study of complex dynamics thereon.

2306.16564

Theodore Zhao

Theodore Zhao, Mu Wei, J. Samuel Preston, Hoifung Poon

Pareto Optimal Learning for Estimating Large Language Model Errors

Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024

10.18653/v1/2024.acl-long.566

cs.CL stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Large Language Models (LLMs) have shown impressive abilities in many applications. When a concrete and precise answer is desired, it is important to have a quantitative estimation of the potential error rate. However, this can be challenging due to the text-in-text-out nature of generative models. We present a method based on Pareto optimization that generates a risk score to estimate the probability of error in an LLM response by integrating multiple sources of information. We prove theoretically that the error estimator optimized in our framework aligns with the LLM and the information sources in an Pareto optimal manner. Experimental results show that the risk scores estimated by our method are well correlated with the true LLM error rate, thus facilitating error correction. By dynamically combining with prompting strategies such as self-verification and information retrieval, we demonstrate the proposed method can be utilized to increase the performance of an LLM, surpassing state-of-the-art task specific models.

2307.01538

Simon Holbach

Simon Holbach, Olivier Raimond

Convergence to the uniform distribution of moderately self-interacting diffusions on compact Riemannian manifolds

math.PR

http://creativecommons.org/licenses/by/4.0/

We consider a self-interacting diffusion $X$ on a smooth compact Riemannian manifold $\mathbb M$, described by the stochastic differential equation \[ dX_t = \sqrt{2} dW_t(X_t)- \beta(t) \nabla V_t(X_t)dt, \] where $\beta$ is suitably lower-bounded and grows at most logarithmically, and $V_t(x)=\frac{1}{t}\int_0^t V(x,X_s)ds$ for a suitable smooth function $V\colon \mathbb M^2\to\mathbb R$ that makes the term $-\nabla V_t(X_t)$ self-repelling. We prove that almost surely the normalized occupation measure $\mu_t$ of $X$ converges weakly to the uniform distribution $\mathcal U$, and we provide a polynomial rate of convergence for smooth test functions. The key to this result is showing that if $f\colon\mathbb M\to\mathbb R$ is smooth, then $\mu_{e^t}(f)$ shadows the flow generated by the ordinary differential equation \[ \dot\nu_t(f)=-\nu_t(f)+\mathcal U(f). \]

2307.04111

Jos\'e Miguel Mateos-Ramos

Jos\'e Miguel Mateos-Ramos, Christian H\"ager, Musa Furkan Keskin, Luc Le Magoarou, Henk Wymeersch

Model-Based End-to-End Learning for Multi-Target Integrated Sensing and Communication under Hardware Impairments

15 pages, 10 figures, accepted to TWC

eess.SP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We study model-based end-to-end learning in the context of integrated sensing and communication (ISAC) under hardware impairments. Hardware impairments are usually addressed by means of array calibration with a focus on communication performance. However, residual impairments may exist that affect sensing performance. This paper proposes a data-driven framework for mitigating such impairments. A monostatic orthogonal frequency-division multiplexing (OFDM) sensing and multiple-input single-output (MISO) communication scenario is considered, incorporating hardware imperfections at the ISAC transceiver antenna array. We propose a novel differentiable version of the orthogonal matching pursuit (OMP) algorithm that is suitable for multi-target sensing and allows for efficient end-to-end learning of the hardware impairments. Based on the differentiable OMP, we devise two model-based parameterization strategies of the ISAC beamformer and sensing receiver to account for hardware impairments: (i) learning a dictionary of steering vectors for different angles and (ii) learning the parameterized hardware impairments. We carry out a comprehensive performance analysis of the proposed model-based learning approaches and a strong baseline consisting of least-squares beamforming, conventional OMP, and maximum-likelihood symbol detection for communication. Results show that by parameterizing the hardware impairments, learning approaches offer gains in terms of higher detection probability, position estimation accuracy, and lower symbol error rate (SER) compared to the baseline. We demonstrate that learning the parameterized hardware impairments outperforms learning a dictionary of steering vectors, also exhibiting the lowest complexity.

2307.04583

Foivos Fioravantes

Foivos Fioravantes, Nikolaos Melissinos, Theofilos Triommatis

Parameterised distance to local irregularity

cs.CC cs.DM cs.DS

http://creativecommons.org/licenses/by/4.0/

A graph $G$ is \emph{locally irregular} if no two of its adjacent vertices have the same degree. In [Fioravantes et al. Complexity of finding maximum locally irregular induced subgraph. {\it SWAT}, 2022], the authors introduced and studied the problem of finding a locally irregular induced subgraph of a given a graph $G$ of maximum order, or, equivalently, computing a subset $S$ of $V(G)$ of minimum order, whose deletion from $G$ results in a locally irregular graph; $S$ is denoted as an \emph{optimal vertex-irregulator of $G$}. In this work we provide an in-depth analysis of the parameterised complexity of computing an optimal vertex-irregulator of a given graph $G$. Moreover, we introduce and study a variation of this problem, where $S$ is a substet of the edges of $G$; in this case, $S$ is denoted as an \emph{optimal edge-irregulator of $G$}. In particular, we prove that computing an optimal vertex-irregulator of a graph $G$ is in FPT when parameterised by the vertex integrity, neighborhood diversity or cluster deletion number of $G$, while it is $W[1]$-hard when parameterised by the feedback vertex set number or the treedepth of $G$. In the case of computing an optimal edge-irregulator of a graph $G$, we prove that this problem is in FPT when parameterised by the vertex integrity of $G$, while it is NP-hard even if $G$ is a planar bipartite graph of maximum degree $4$, and $W[1]$-hard when parameterised by the size of the solution, the feedback vertex set or the treedepth of $G$. Our results paint a comprehensive picture of the tractability of both problems studied here, considering most of the standard graph-structural parameters.

2307.05219

David Rapado-Rincon

David Rapado-Rincon, Eldert J. van Henten, Gert Kootstra

MinkSORT: A 3D deep feature extractor using sparse convolutions to improve 3D multi-object tracking in greenhouse tomato plants

10.1016/j.biosystemseng.2023.11.003

cs.RO

http://creativecommons.org/licenses/by/4.0/

The agro-food industry is turning to robots to address the challenge of labour shortage. However, agro-food environments pose difficulties for robots due to high variation and occlusions. In the presence of these challenges, accurate world models, with information about object location, shape, and properties, are crucial for robots to perform tasks accurately. Building such models is challenging due to the complex and unique nature of agro-food environments, and errors in the model can lead to task execution issues. In this paper, MinkSORT, a novel method for generating tracking features using a 3D sparse convolutional network in a deepSORT-like approach, is proposed to improve the accuracy of world models in agro-food environments. MinkSORT was evaluated using real-world data collected in a tomato greenhouse, where it significantly improved the performance of a baseline model that tracks tomato positions in 3D using a Kalman filter and Mahalanobis distance. MinkSORT improved the HOTA from 42.8% to 44.77%, the association accuracy from 32.55% to 35.55%, and the MOTA from 57.63% to 58.81%. Different contrastive loss functions for training MinkSORT were also evaluated, and it was demonstrated that it leads to improved performance in terms of three separate precision and recall detection outcomes. The proposed method improves world model accuracy, enabling robots to perform tasks such as harvesting and plant maintenance with greater efficiency and accuracy, which is essential for meeting the growing demand for food in a sustainable manner.

2307.07780

Olga Mula

Wolfgang Dahmen and Olga Mula

Accuracy Controlled Schemes for the Eigenvalue Problem of the Radiative Transfer Equation

math.NA cs.NA math.AP math.SP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The criticality problem in nuclear engineering asks for the principal eigenpair of a Boltzmann operator describing neutron transport in a reactor core. Being able to reliably design, and control such reactors requires assessing these quantities within quantifiable accuracy tolerances. In this paper we propose a paradigm that deviates from the common practice of approximately solving the corresponding spectral problem with a fixed, presumably sufficiently fine discretization. Instead, the present approach is based on first contriving iterative schemes, formulated in function space, that are shown to converge at a quantitative rate without assuming any a priori excess regularity properties, and that exploit only properties of the optical parameters in the underlying radiative transfer model. We develop the analytical and numerical tools for approximately realizing each iteration step within judiciously chosen accuracy tolerances, verified by a posteriori estimates, so as to still warrant quantifiable convergence to the exact eigenpair. This is carried out in full first for a Newton scheme. Since this is only locally convergent we analyze in addition the convergence of a power iteration in function space to produce sufficiently accurate initial guesses. Here we have to deal with intrinsic difficulties posed by compact but unsymmetric operators preventing standard arguments used in the finite dimensional case. Our main point is that we can avoid any condition on an initial guess to be already in a small neighborhood of the exact solution. We close with a discussion of remaining intrinsic obstructions to a certifiable numerical implementation, mainly related to not knowing the gap between the principal eigenvalue and the next smaller one in modulus.

2307.08310

Gerrit Schierholz

Gerrit Schierholz

Repercussions of the Peccei-Quinn axion on QCD

11 pages, expanded by new findings, including two figures; matches published version

DESY-23-101

hep-ph hep-lat hep-th nucl-th

http://creativecommons.org/licenses/by/4.0/

The axion, originally postulated by Peccei and Quinn to solve the strong CP problem, has become of great interest in particle and astroparticle phenomenology. Yet it has a problem. It is widely assumed that the axion leaves the nonperturbative features of QCD unscathed. This is, however, not the case. On the contrary, the axion extension is found to be in conflict with seminal results on the low-energy properties of QCD. The key is that the anomalous coupling of the axion to the gauge bosons can be integrated out, leaving behind a path integral over topologically trivial gauge potentials.

2307.09423

Jens Tuyls

Jens Tuyls, Dhruv Madeka, Kari Torkkola, Dean Foster, Karthik Narasimhan, Sham Kakade

Scaling Laws for Imitation Learning in Single-Agent Games

Accepted at TMLR 2024

cs.LG cs.AI stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Imitation Learning (IL) is one of the most widely used methods in machine learning. Yet, many works find it is often unable to fully recover the underlying expert behavior, even in constrained environments like single-agent games. However, none of these works deeply investigate the role of scaling up the model and data size. Inspired by recent work in Natural Language Processing (NLP) where "scaling up" has resulted in increasingly more capable LLMs, we investigate whether carefully scaling up model and data size can bring similar improvements in the imitation learning setting for single-agent games. We first demonstrate our findings on a variety of Atari games, and thereafter focus on the extremely challenging game of NetHack. In all games, we find that IL loss and mean return scale smoothly with the compute budget (FLOPs) and are strongly correlated, resulting in power laws for training compute-optimal IL agents. Finally, we forecast and train several NetHack agents with IL and find they outperform prior state-of-the-art by 1.5x in all settings. Our work both demonstrates the scaling behavior of imitation learning in a variety of single-agent games, as well as the viability of scaling up current approaches for increasingly capable agents in NetHack, a game that remains elusively hard for current AI systems.

2307.11002

Tobias Lenz

Tobias Lenz, Anna Marie Schr\"oter

Simplicial $*$-modules and mild actions

Minor corrections and improvements following a referee report. 27 pages

Homology Homotopy Appl. 26 No. 2 (2024), pp. 229-258

10.4310/HHA.2024.v26.n2.a12

math.AT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We develop an analogue of the theory of $*$-modules in the world of simplicial sets, based on actions of a certain simplicial monoid $E\mathcal M$ originally appearing in the construction of global algebraic $K$-theory. As our main results, we show that strictly commutative monoids with respect to a certain box product on these simplicial $*$-modules yield models of equivariantly and globally coherently commutative monoids, and we give a characterization of simplicial $*$-modules in terms of a certain mildness condition on the $E\mathcal M$-action, relaxing the notion of tameness previously investigated by Sagave-Schwede and the first author.

2307.13442

Jingpeng Wu

Jingpeng Wu

The gyrokinetic limit for the two dimensional Vlasov-Poisson system with multi-point charges

34 pages; typos corrected

math.AP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this article, we investigate the gyrokinetic limit for the two dimensional Vlasov-Poisson system with multi-point charges. We show that the solution converges to a measure-valued solution of the Euler equation with a defect measure, which extends the results of Miot (Nonlinearity 32(2):654-677, 2019) to the case of multi-point charges and removes the smallness condition $\sup_{0<\varepsilon<1}\|f_{\varepsilon}^0\|_{L^1}<1$. The main difficulty arises from the fact that the orbits of point charges will intersect frequently and rapidly as the magnetic field intensity becomes large. To overcome the problem, we adopt the techniques recently developed by Wu and Zhang (Journal of Statistical Physics 190:183, 2023) combined with a new technique introduced here.

2307.16519

Felix Fie{\ss}inger

Felix Fie{\ss}inger and Mitja Stadje

The $C^{0,1}$ It\^o-Ventzell formula for weak Dirichlet processes

math.PR

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper proves an extension of the It\^o-Ventzell formula that applies to stochastic flows in $C^{0,1}$ for continuous weak Dirichlet processes. We apply this theorem, for example, to give a representation result for strong solutions of time-dependent elliptic SPDEs, to derive formulas for quadratic variations, and to relax assumptions in a financial mathematics context.

2308.03240

Xin Chen

Xin Chen, Andy Sun, Wenbo Shi, Na Li

Carbon-Aware Optimal Power Flow

math.OC cs.SY eess.SY

http://creativecommons.org/licenses/by-nc-nd/4.0/

To facilitate effective decarbonization of the electric power sector, this paper introduces the generic Carbon-aware Optimal Power Flow (C-OPF) method for power system decision-making that considers demand-side carbon accounting and emission management. Built upon the classic optimal power flow (OPF) model, the C-OPF method incorporates carbon emission flow equations and constraints, as well as carbon-related objectives, to jointly optimize power flow and carbon flow. In particular, this paper establishes the feasibility and solution uniqueness of the carbon emission flow equations, and proposes modeling and linearization techniques to address the issues of undetermined power flow directions and bilinear terms in the C-OPF model. Additionally, two novel carbon emission models, together with the carbon accounting schemes, for energy storage systems are developed and integrated into the C-OPF model. Numerical simulations demonstrate the characteristics and effectiveness of the C-OPF method, in comparison with OPF solutions.

2308.04167

Naomi Schneider

Naomi Schneider, Volker Michel, Nico Sneeuw

High-dimensional experiments for the downward continuation using the LRFMP algorithm

GEM - International Journal on Geomathematics, Volume 16, article number 4 (2025)

10.1007/s13137-024-00255-y

math.NA cs.NA

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Time-dependent gravity data from satellite missions like GRACE-FO reveal mass redistribution in the system Earth at various time scales: long-term climate change signals, inter-annual phenomena like El Nino, seasonal mass transports and transients, e. g. due to earthquakes. For this contemporary issue, a classical inverse problem has to be considered: the gravitational potential has to be modelled on the Earth's surface from measurements in space. This is also known as the downward continuation problem. Thus, it is important to further develop current mathematical methods for such inverse problems. For this, the (Learning) Inverse Problem Matching Pursuits ((L)IPMPs) have been developed within the last decade. Their unique feature is the combination of local as well as global trial functions in the approximative solution of an inverse problem such as the downward continuation of the gravitational potential. In this way, they harmonize the ideas of a traditional spherical harmonic ansatz and the radial basis function approach. Previous publications on these methods showed proofs of concept. Here, we consider the methods for high-dimensional experiments settings with more than 500 000 grid points which yields a resolution of 20 km at best on a realistic satellite geometry. We also explain the changes in the methods that had to be done to work with such a large amount of data. The corresponding code (updated for big data use) is available at https://doi.org/10.5281/zenodo.8223771 under the licence CC BY-NC-SA 3.0 Germany.

2308.05203

Zhiyuan Wang

Zhiyuan Wang and Kaden R. A. Hazzard

Particle exchange statistics beyond fermions and bosons

27 pages, Mathematica codes accompanying this paper are available at https://github.com/lagrenge94/Mathematica-codes-for-parastatistics

quant-ph cond-mat.stat-mech cond-mat.str-el hep-th math-ph math.MP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

It is commonly believed that there are only two types of particle exchange statistics in quantum mechanics, fermions and bosons, with the exception of anyons in two dimension. In principle, a second exception known as parastatistics, which extends outside of two dimensions, has been considered but was believed to be physically equivalent to fermions and bosons. In this paper we show that nontrivial parastatistics inequivalent to either fermions or bosons can exist in physical systems. These new types of identical particles obey generalized exclusion principles, leading to exotic free-particle thermodynamics distinct from any system of free fermions and bosons. We formulate our theory by developing a second quantization of paraparticles, which naturally includes exactly solvable non-interacting theories, and incorporates physical constraints such as locality. We then construct a family of exactly solvable quantum spin models in one and two dimensions where free paraparticles emerge as quasiparticle excitations, and their exchange statistics can be physically observed and is notably distinct from fermions and bosons. This demonstrates the possibility of a new type of quasiparticle in condensed matter systems, and, more speculatively, the potential for previously unconsidered types of elementary particles.

2308.05246

Isamiddin Rakhimov Prof.

D.Asrorov, U.Bekbaev, I.Rakhimov

Classification of Frobenius algebra structures on two-dimensional vector space over any base field

math.RA

http://creativecommons.org/licenses/by/4.0/

Classifying Frobenius algebras is a key question that has been addressed in various contexts. The structure of finite-dimensional Frobenius algebras depends on the base field and the dimension of the algebra, leading to different classification results depending on whether the base field has characteristic zero, characteristic $p$, or other properties. Frobenius algebras over fields of characteristic zero have been well-studied, often related to semisimple algebra theory. The behavior of Frobenius algebras over fields of positive characteristic presents new challenges, with connections to modular representation theory. In the paper, we first classify all associative algebra structures on a two-dimensional vector space over any base field equipped with a non-degenerate bilinear form. We then identify which of these are Frobenius algebras. Lists of canonical representatives of the isomorphism classes of these algebras are provided for a base field with characteristics not equal to two or three, as well as for characteristics two and three.

2308.06706

Ole Steuernagel

Ole Steuernagel and Ray-Kuang Lee

Wigner's Phase Space Current for Variable Beam Splitters -Seeing Beam Splitters in a New Light-

5 figures. arXiv admin note: text overlap with arXiv:2307.16510

10.1364/JOSAB.545875

quant-ph

http://creativecommons.org/licenses/by-nc-nd/4.0/

Beam splitters allow us to superpose two continuous single mode quantum systems. To study the behaviour of their strongly mode mixing dynamics we consider variable beam splitters and their dynamics using Wigner's phase space distribution, W, the evolution of which is governed by the continuity-equation $ \frac{\partial}{\partial \tau} W = - {\nabla } \cdot {J}$. We derive the form of the corresponding Wigner current, J, of each outgoing mode after tracing out the other. The influence of the modes on each other is analyzed and visualized using their respective Wigner distributions and Wigner currents. This allows us to perform geometrical analyses of the mode interactions, casting new light on beam splitter behaviour. Several of the presented results should be immediately testable in experiments.

2308.09524

Flavio Lorez

Flavio Lorez, Mohit Pundir, David S. Kammer

Eulerian framework for contact between solids represented as phase fields

10.1016/j.cma.2023.116497

cond-mat.soft

http://creativecommons.org/licenses/by/4.0/

Mechanical contact between solids is almost exclusively modeled in Lagrangian frameworks. While these frameworks have been developed extensively and applied successfully to numerous contact problems, they generally require complex algorithms for contact detection and resolution. These challenges become particularly important when contact appears between solids with evolving boundaries, such as in systems where crystals grow in a constrained space. In this work, we introduce a fully Eulerian finite element framework for modeling contact between elastic solids tailored towards problems including evolving and intricate surfaces. The proposed approach uses a phase-field method that involves a diffuse representation of geometries on a fixed mesh, simplifying the modeling of evolving surfaces. Our methodology introduces a novel volumetric contact constraint based on penalty body forces, efficiently resolving the interpenetration of solids. We showcase the validity and versatility of our method through numerical examples, highlighting its ability to accurately capture complex solid-solid interactions. The Eulerian phase-field formulation greatly simplifies contact detection and its resolution. Furthermore, the framework can be straightforwardly coupled with other physical phenomena through the inclusion of multiple energy terms in the evolution of the phase-field. This enables multiphysics modeling, potentially providing a valuable tool for a wide range of applications involving chemically or physically evolving deformable solids in contact, as they commonly occur in deterioration processes of porous media.

2308.15084

Michele Tucci

Vittorio Cortellessa, J. Andres Diaz-Pace, Daniele Di Pompeo, Sebastian Frank, Pooyan Jamshidi, Michele Tucci, Andr\'e van Hoorn

Introducing Interactions in Multi-Objective Optimization of Software Architectures

cs.SE

http://creativecommons.org/licenses/by/4.0/

Software architecture optimization aims to enhance non-functional attributes like performance and reliability while meeting functional requirements. Multi-objective optimization employs metaheuristic search techniques, such as genetic algorithms, to explore feasible architectural changes and propose alternatives to designers. However, this resource-intensive process may not always align with practical constraints. This study investigates the impact of designer interactions on multi-objective software architecture optimization. Designers can intervene at intermediate points in the fully automated optimization process, making choices that guide exploration towards more desirable solutions. Through several controlled experiments as well as an initial user study (14 subjects), we compare this interactive approach with a fully automated optimization process, which serves as a baseline. The findings demonstrate that designer interactions lead to a more focused solution space, resulting in improved architectural quality. By directing the search towards regions of interest, the interaction uncovers architectures that remain unexplored in the fully automated process. In the user study, participants found that our interactive approach provides a better trade-off between sufficient exploration of the solution space and the required computation time.

2309.02495

Djuna Croon

Djuna Croon, and Juri Smirnov

Dark Matter (H)eats Young Planets

8 pages, 4 figures, includes astronomical observations made with the naked eye. V2: improved analysis, updated constraints. V3: version published in JCAP

JCAP 11 (2024), 046

10.1088/1475-7516/2024/11/046

IPPP/23/43, LTH-1349

hep-ph astro-ph.EP astro-ph.HE

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We study the effect of dark matter annihilation on the formation of Jovian planets. We show that dark matter heat injections can slow or halt Kelvin-Helmholtz contraction, preventing the accretion of hydrogen and helium onto the solid core. The existence of Jupiter in our solar system can therefore be used to infer constraints on dark matter with relatively strong interaction cross sections. We derive novel constraints on the cross section for both spin-dependent and spin-independent dark matter. We highlight the possibility of a positive detection using future observations by JWST, which could reveal strongly varying planet morpholoiges close to our Galactic Center.

2309.03579

Ajitesh Srivastava

Ajitesh Srivastava

DTW+S: Shape-based Comparison of Time-series with Ordered Local Trend

Longer version of the paper "Aligning Time-series by Local Trends: Applications in Public Health" accepted at The 39th Annual AAAI Conference on Artificial Intelligence (AAAI 2025)

cs.LG cs.AI

http://creativecommons.org/licenses/by/4.0/

Measuring distance or similarity between time-series data is a fundamental aspect of many applications including classification, clustering, and ensembling/alignment. Existing measures may fail to capture similarities among local trends (shapes) and may even produce misleading results. Our goal is to develop a measure that looks for similar trends occurring around similar times and is easily interpretable for researchers in applied domains. This is particularly useful for applications where time-series have a sequence of meaningful local trends that are ordered, such as in epidemics (a surge to an increase to a peak to a decrease). We propose a novel measure, DTW+S, which creates an interpretable "closeness-preserving" matrix representation of the time-series, where each column represents local trends, and then it applies Dynamic Time Warping to compute distances between these matrices. We present a theoretical analysis that supports the choice of this representation. We demonstrate the utility of DTW+S in several tasks. For the clustering of epidemic curves, we show that DTW+S is the only measure able to produce good clustering compared to the baselines. For ensemble building, we propose a combination of DTW+S and barycenter averaging that results in the best preservation of characteristics of the underlying trajectories. We also demonstrate that our approach results in better classification compared to Dynamic Time Warping for a class of datasets, particularly when local trends rather than scale play a decisive role.

2309.04700

Phuong Duy Huynh Mr.

Phuong Duy Huynh, Thisal De Silva, Son Hoang Dau, Xiaodong Li, Iqbal Gondal, Emanuele Viterbo

From Programming Bugs to Multimillion-Dollar Scams: An Analysis of Trapdoor Tokens on Uniswap

22 pages, 11 figures

cs.CR

http://creativecommons.org/licenses/by/4.0/

We investigate in this work a recently emerged type of scam ERC-20 token called Trapdoor, which has cost investors billions of US dollars on Uniswap, the largest decentralised exchange on Ethereum, from 2020 to 2023. In essence, Trapdoor tokens allow users to buy but preventing them from selling by embedding logical bugs and/or owner-only features in their smart contracts. By manually inspecting a number of Trapdoor samples, we established the first systematic classification of Trapdoor tokens and a comprehensive list of techniques that scammers used to embed and conceal malicious codes, accompanied by a detailed analysis of representative scam contracts. In particular, we developed TrapdoorAnalyser, a fine-grained detection tool that generates and crosschecks the error-log of a buy-and-sell test and the list of embedded Trapdoor indicators from a contract-semantic check to reliably identify a Trapdoor token. TrapdoorAnalyser not only outperforms the state-of-the-art commercial tool GoPlus in accuracy, but also provides traces of malicious code with a full explanation, which most of the existing tools lack. Using TrapdoorAnalyser, we constructed the very first dataset of about 30,000 Trapdoor and non-Trapdoor tokens on UniswapV2, which allows us to train several machine learning algorithms that can detect with very high accuracy even Trapdoor tokens with no available Solidity source codes.

2309.06287

David Bevan

David Bevan and Dan Threlfall

On the evolution of random integer compositions

39 pages; to appear in the Electronic Journal of Combinatorics

math.CO

http://creativecommons.org/licenses/by/4.0/

We explore how the asymptotic structure of a random $n$-term weak integer composition of $m$ evolves, as $m$ increases from zero. The primary focus is on establishing thresholds for the appearance and disappearance of substructures. These include the longest and shortest runs of zero terms or of nonzero terms, longest increasing runs, longest runs of equal terms, largest squares (runs of $k$ terms each equal to $k$), as well as a wide variety of other patterns. Of particular note is the dichotomy between the appearance and disappearance of exact consecutive patterns, with smaller patterns appearing before larger ones, whereas longer patterns disappear before shorter ones.

2309.07520

K Ashok Kumar

K Ashok Kumar and Nirjan Biswas

Strict Faber-Krahn type inequality for the mixed local-nonlocal operator under polarization

13 pages. This article significantly expands upon the concepts presented in our previous paper, arXiv:2305.16672, to address a broader set of cases

math.AP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Let $\Omega \subset \mathbb{R}^d$ with $d\geq 2$ be a bounded domain of class $\mathcal{C}^{1,\beta }$ for some $\beta \in (0,1)$. For $p\in (1, \infty )$ and $s\in (0,1)$, let $\Lambda ^s_{p}(\Omega )$ be the first eigenvalue of the mixed local-nonlocal operator $-\Delta _p+(-\Delta _p)^s$ in $\Omega $ with the homogeneous nonlocal Dirichlet boundary condition. We establish a strict Faber-Krahn type inequality for $\Lambda _{p}^s(\cdot )$ under polarization. As an application of this strict inequality, we obtain the strict monotonicity of $\Lambda _{p}^s(\cdot )$ over annular domains and characterize the rigidity property of the balls in the classical Faber-Krahn inequality for $-\Delta _p+(-\Delta _p)^s$.

2309.07675

Mehdi Zadem

Mehdi Zadem and Sergio Mover and Sao Mai Nguyen

Goal Space Abstraction in Hierarchical Reinforcement Learning via Set-Based Reachability Analysis

10.1109/ICDL55364.2023.10364473

cs.LG cs.AI

http://creativecommons.org/licenses/by-nc-nd/4.0/

Open-ended learning benefits immensely from the use of symbolic methods for goal representation as they offer ways to structure knowledge for efficient and transferable learning. However, the existing Hierarchical Reinforcement Learning (HRL) approaches relying on symbolic reasoning are often limited as they require a manual goal representation. The challenge in autonomously discovering a symbolic goal representation is that it must preserve critical information, such as the environment dynamics. In this paper, we propose a developmental mechanism for goal discovery via an emergent representation that abstracts (i.e., groups together) sets of environment states that have similar roles in the task. We introduce a Feudal HRL algorithm that concurrently learns both the goal representation and a hierarchical policy. The algorithm uses symbolic reachability analysis for neural networks to approximate the transition relation among sets of states and to refine the goal representation. We evaluate our approach on complex navigation tasks, showing the learned representation is interpretable, transferrable and results in data efficient learning.

2309.08846

Felipe Flores Llarena

Felipe I. Flores

Twisted convolution algebras with coefficients on a differential subalgebra

9 pages. Contains changes in redaction, specially in the introduction. To appear in Studia Mathematica

math.OA

http://creativecommons.org/licenses/by/4.0/

Let $({\sf G},\alpha, \omega,\mathfrak B)$ be a measurable twisted action of the locally compact group ${\sf G}$ on a Banach $^*$-algebra $\mathfrak B$ and $\mathfrak A$ a differential Banach $^*$-subalgebra of $\mathfrak B$, which is stable under said action. We observe that $L^1_{\alpha,\omega}({\sf G},\mathfrak A)$ is a differential subalgebra of $L^1_{\alpha,\omega}({\sf G},\mathfrak B)$. We use this fact to provide new examples of groups with symmetric Banach $^*$-algebras. In particular, we prove that discrete rigidly symmetric extensions of compact groups are symmetric or that semidirect products ${\sf K}\rtimes{\sf H}$, with ${\sf H}$ symmetric and ${\sf K}$ compact, are symmetric.

2309.10230

Shaocong Xu

Shaocong Xu, Pengfei Li, Qianpu Sun, Xinyu Liu, Yang Li, Shihui Guo, Zhen Wang, Bo Jiang, Rui Wang, Kehua Sheng, Bo Zhang, Li Jiang, Hao Zhao, and Yilun Chen

LiON: Learning Point-wise Abstaining Penalty for LiDAR Outlier DetectioN Using Diverse Synthetic Data

Accepted by AAAI2025. Codes are available at https://github.com/Daniellli/LiON/

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

LiDAR-based semantic scene understanding is an important module in the modern autonomous driving perception stack. However, identifying outlier points in a LiDAR point cloud is challenging as LiDAR point clouds lack semantically-rich information. While former SOTA methods adopt heuristic architectures, we revisit this problem from the perspective of Selective Classification, which introduces a selective function into the standard closed-set classification setup. Our solution is built upon the basic idea of abstaining from choosing any inlier categories but learns a point-wise abstaining penalty with a margin-based loss. Apart from learning paradigms, synthesizing outliers to approximate unlimited real outliers is also critical, so we propose a strong synthesis pipeline that generates outliers originated from various factors: object categories, sampling patterns and sizes. We demonstrate that learning different abstaining penalties, apart from point-wise penalty, for different types of (synthesized) outliers can further improve the performance. We benchmark our method on SemanticKITTI and nuScenes and achieve SOTA results. Codes are available at https://github.com/Daniellli/LiON/.

2309.10978

Sean Mann

Sean Mann

Negative Spillover: A Potential Source of Bias in Pragmatic Clinical Trials

6.5 pages of main text, 2 figures, 1 table; New version with title change and minor edits to main text

Trials 25, 833 (2024)

10.1186/s13063-024-08675-9

stat.ME q-bio.QM

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Pragmatic clinical trials evaluate the effectiveness of health interventions in real-world settings. Negative spillover can arise in a pragmatic trial if the study intervention affects how scarce resources are allocated between patients in the intervention and comparison groups. This can harm patients assigned to the control group and lead to overestimation of treatment effect. While this type of negative spillover is often addressed in trials of social welfare and public health interventions, there is little recognition of this source of bias in the medical literature. In this article, I examine what causes negative spillover and how it may have led clinical trial investigators to overestimate the effect of patient navigation, AI-based physiological alarms, and elective induction of labor. I also suggest ways to detect negative spillover and design trials that avoid this potential source of bias.

2309.11036

Tianyi Han

Tianyi Han, Jingya Li, Zhipeng Guo and Yuan Jin

Scalable Acceleration for Classification-Based Derivative-Free Optimization

cs.LG cs.NA math.NA math.OC

http://creativecommons.org/licenses/by-nc-nd/4.0/

Derivative-free optimization algorithms play an important role in scientific and engineering design optimization problems, especially when derivative information is not accessible. In this paper, we study the framework of sequential classification-based derivative-free optimization algorithms. By introducing learning theoretic concept hypothesis-target shattering rate, we revisit the computational complexity upper bound of SRACOS (Hu, Qian, and Yu 2017). Inspired by the revisited upper bound, we propose an algorithm named RACE-CARS, which adds a random region-shrinking step compared with SRACOS. We further establish theorems showing the acceleration by region shrinking. Experiments on the synthetic functions as well as black-box tuning for language-model-as-a-service demonstrate empirically the efficiency of RACE-CARS. An ablation experiment on the introduced hyperparameters is also conducted, revealing the mechanism of RACE-CARS and putting forward an empirical hyper-parameter tuning guidance.

2309.14598

Li Chen

Guozhong Zheng, Jiqiang Zhang, Jing Zhang, Weiran Cai, and Li Chen

Decoding trust: A reinforcement learning perspective

12 pages, 11 figures. Comments are appreciated

New Journal of Physics 26, 053041 (2024)

10.1088/1367-2630/ad4b5a

q-bio.PE cond-mat.dis-nn nlin.AO stat.ML

http://creativecommons.org/licenses/by/4.0/

Behavioral experiments on the trust game have shown that trust and trustworthiness are universal among human beings, contradicting the prediction by assuming \emph{Homo economicus} in orthodox Economics. This means some mechanism must be at work that favors their emergence. Most previous explanations however need to resort to some factors based upon imitative learning, a simple version of social learning. Here, we turn to the paradigm of reinforcement learning, where individuals update their strategies by evaluating the long-term return through accumulated experience. Specifically, we investigate the trust game with the Q-learning algorithm, where each participant is associated with two evolving Q-tables that guide one's decision making as trustor and trustee respectively. In the pairwise scenario, we reveal that high levels of trust and trustworthiness emerge when individuals appreciate both their historical experience and returns in the future. Mechanistically, the evolution of the Q-tables shows a crossover that resembles human's psychological changes. We also provide the phase diagram for the game parameters, where the boundary analysis is conducted. These findings are robust when the scenario is extended to a latticed population. Our results thus provide a natural explanation for the emergence of trust and trustworthiness without external factors involved. More importantly, the proposed paradigm shows the potential in deciphering many puzzles in human behaviors.

2309.15989

Kirsty Scott

Kirsty Scott, Elliot Kisiel, Flora Yakhou, Stefano Agrestini, Mirian Garcia-Fernandez, Kurt Kummer, Jaewon Choi, Ruidan Zhong, John A. Schneeloch, Genda D. Gu, Ke-Jin Zhou, Nicholas B. Brookes, Alexander F. Kemper, Matteo Minola, Fabio Boschini, Alex Frano, Adrian Gozar, and Eduardo H. da Silva Neto

Detection of a two-phonon mode in a cuprate superconductor via polarimetric RIXS

10.1103/PhysRevB.109.125126

cond-mat.str-el cond-mat.supr-con

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Recent improvements in the energy resolution of resonant inelastic x-ray scattering experiments (RIXS) at the Cu-L$_3$ edge have enabled the study of lattice, spin, and charge excitations. Here, we report on the detection of a low intensity signal at 140meV, twice the energy of the bond-stretching (BS) phonon mode, in the cuprate superconductor $\textrm{Bi}_2\textrm{Sr}_2\textrm{Ca}\textrm{Cu}_2\textrm{O}_{8+x}$ (Bi-2212). Ultra-high resolution polarimetric RIXS measurements allow us to resolve the outgoing polarization of the signal and identify this feature as a two-phonon excitation. Further, we study the connection between the two-phonon mode and the BS one-phonon mode by constructing a joint density of states toy model that reproduces the key features of the data.

2309.16120

Zhao Tian

Zhao Tian, Junjie Chen, Xiangyu Zhang

Fixing Large Language Models' Specification Misunderstanding for Better Code Generation

Accepted by ICSE 2025

cs.SE

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Code generation is to automatically generate source code conforming to a given programming specification, which has received extensive attention especially with the development of large language models (LLMs). Due to the inherent difficulty of code generation, the code generated by LLMs may not be aligned with the specification. Although thought-eliciting prompting techniques have been proposed to enhance the code generation performance of LLMs, producing correct understanding for complicated programming problems remains challenging, resulting in unsatisfactory performance. Also, some feedback-based prompting techniques have been proposed to fix incorrect code using error messages produced by test execution. However, when the generated code deviates significantly from the ground truth, they encounter difficulties in improving performance based on such coarse-grained information. In this work, we propose a novel prompting technique, called {\mu}FiX, to improve the code generation performance of LLMs by devising both sophisticated thought-eliciting prompting and feedback-based prompting and making the first exploration on their synergy. It first exploits test case analysis to obtain specification understanding and enables a self-improvement process to identify and refine the misunderstanding in the thought-eliciting prompting phase. {\mu}FiX further fixes the specification understanding towards the direction reducing the gap between the provided understanding (from the first phase) and the actual understanding implicitly utilized by LLMs for code generation in the feedback-based prompting phase. By improving the understanding with {\mu}FiX, the code generation performance of LLMs can be largely improved. Our evaluation on two advanced LLMs (ChatGPT and DeepSeek-Coder) with six widely-used benchmarks by comparing with 15 baselines, demonstrates the effectiveness of {\mu}FiX ...

2309.16590

Marco Barbieri

Marco Barbieri, and Primo\v{z} Poto\v{c}nik

Vertex-primitive digraphs with large fixity

16 pages

Annali di Matematica Pura ed Applicata, 203: 2383-2403, (2024)

10.1007/s10231-024-01447-x

math.CO math.GR

http://creativecommons.org/licenses/by/4.0/

The relative fixity of a digraph $\Gamma$ is defined as the ratio between the largest number of vertices fixed by a nontrivial automorphism of $\Gamma$ and the number of vertices of $\Gamma$. We characterize the vertex-primitive digraphs whose relative fixity is at least $1/3$, and we show that there are only finitely many vertex-primitive digraphs of bounded out-valency and relative fixity exceeding a positive constant.

2309.16919

Yu-Kuo Hsiao

Yu-Kuo Hsiao

Study of two-body doubly charmful baryonic $B$ decays with $SU(3)$ flavor symmetry

11 pages, 1 figure, 2 tables, updated to match published version

JHEP11, 117 (2023)

10.1007/JHEP11(2023)117

hep-ph hep-ex

http://creativecommons.org/licenses/by/4.0/

Within the framework of $SU(3)$ flavor symmetry, we investigate two-body doubly charmful baryonic $B\to{\bf B}_c\bar{\bf B}'_c$ decays, where ${\bf B}_c\bar{\bf B}'_c$ represents the anti-triplet charmed dibaryon. We determine the $SU(3)_f$ amplitudes and calculate ${\cal B}(B^-\to \Xi_c^0\bar \Xi_c^-)=(3.4^{+1.0}_{-0.9})\times 10^{-5}$ and ${\cal B}(\bar B^0_s\to \Lambda_c^+\bar \Xi_c^-)=(3.9^{+1.2}_{-1.0})\times 10^{-5}$ induced by the single $W$-emission configuration. We find that the $W$-exchange amplitude, previously neglected in studies, needs to be taken into account. It can cause a destructive interfering effect with the $W$-emission amplitude, alleviating the significant discrepancy between the theoretical estimation and experimental data for ${\cal B}(\bar B^0\to\Lambda_c^+\bar\Lambda_c^-)$. To test other interfering decay channels, we calculate ${\cal B}(\bar B^0_s\to \Xi_c^{0(+)}\bar \Xi_c^{0(+)})=(3.0^{+1.4}_{-1.1})\times 10^{-4}$ and ${\cal B}(\bar B^0\to \Xi_c^0\bar \Xi_c^0)=(1.5^{+0.7}_{-0.6})\times 10^{-5}$. We estimate non-zero branching fractions for the pure $W$-exchange decay channels, specifically ${\cal B}(\bar B^0_s\to \Lambda_c^+\bar \Lambda_c^-)=(8.1^{+1.7}_{-1.5})\times 10^{-5}$ and ${\cal B}(\bar B^0\to \Xi_c^+\bar \Xi_c^-)=(3.0\pm 0.6)\times 10^{-6}$. Additionally, we predict ${\cal B}(B^+_c\to \Xi_c^+\bar \Xi_c^0)=(2.8^{+0.9}_{-0.7})\times 10^{-4}$ and ${\cal B}(B^+_c\to \Lambda_c^+\bar \Xi_c^0)=(1.6^{+0.5}_{-0.4})\times 10^{-5}$, which are accessible to experimental facilities such as LHCb.

2310.02686

Alessio Paviglianiti

Alessio Paviglianiti and Xhek Turkeshi and Marco Schir\`o and Alessandro Silva

Enhanced Entanglement in the Measurement-Altered Quantum Ising Chain

17 pages, 5 figures; updated to the version accepted by Quantum

quant-ph cond-mat.stat-mech

http://creativecommons.org/licenses/by/4.0/

Understanding the influence of measurements on the properties of many-body systems is a fundamental problem in quantum mechanics and for quantum technologies. This paper explores how a finite density of stochastic local measurement modifies a given state's entanglement structure. Considering various measurement protocols, we explore the typical quantum correlations of their associated projected ensembles arising from the ground state of the quantum Ising model. Using large-scale numerical simulations, we demonstrate substantial differences among inequivalent measurement protocols. Surprisingly, we observe that forced on-site measurements can enhance both bipartite and multipartite entanglement. We present a phenomenological toy model and perturbative calculations to analytically support these results. Furthermore, we extend these considerations to the non-Hermitian Ising model, naturally arising in optically monitored systems, and we show that its qualitative entanglement features are not altered by a finite density of projective measurements. Overall, these results reveal a complex phenomenology where local quantum measurements do not simply disentangle degrees of freedom, but may actually strengthen the entanglement in the system.

2310.03493

Felix Finster

Felix Finster, Magdalena Lottner and Alexander V. Sobolev

The Fermionic Entanglement Entropy and Area Law for the Relativistic Dirac Vacuum State

37 pages, LaTeX, a few typos corrected (published version)

Adv.Theor.Math.Phys. 28 (2024) 1933-1985

10.4310/ATMP.241031011130

math-ph math.MP math.SP quant-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We consider the fermionic entanglement entropy for the free Dirac field in a bounded spatial region of Minkowski spacetime. In order to make the system ultraviolet finite, a regularization is introduced. An area law is proven in the limiting cases where the volume tends to infinity and/or the regularization length tends to zero. The technical core of the paper is to generalize a theorem of Harold Widom to pseudo-differential operators whose principal symbols develop a specific discontinuity at a single point.

2310.07926

Joseph Slote

Lars Becker, Ohad Klein, Joseph Slote, Alexander Volberg, Haonan Zhang

Dimension-free discretizations of the uniform norm by small product sets

27 pages. Previously titled, "Dimension-free Remez Inequalities and norm designs."

Invent. Math. (2024)

10.1007/s00222-024-01306-9

math.FA math.AP math.CA math.PR quant-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Let $f$ be an analytic polynomial of degree at most $K-1$. A classical inequality of Bernstein compares the supremum norm of $f$ over the unit circle to its supremum norm over the sampling set of the $K$-th roots of unity. Many extensions of this inequality exist, often understood under the umbrella of Marcinkiewicz-Zygmund-type inequalities for $L^p,1\le p\leq \infty$ norms. We study dimension-free extensions of these discretization inequalities in the high-dimension regime, where existing results construct sampling sets with cardinality growing with the total degree of the polynomial. In this work we show that dimension-free discretizations are possible with sampling sets whose cardinality is independent of $\deg(f)$ and is instead governed by the maximum individual degree of $f$; i.e., the largest degree of $f$ when viewed as a univariate polynomial in any coordinate. For example, we find that for $n$-variate analytic polynomials $f$ of degree at most $d$ and individual degree at most $K-1$, $\|f\|_{L^\infty(\mathbf{D}^n)}\leq C(X)^d\|f\|_{L^\infty(X^n)}$ for any fixed $X$ in the unit disc $\mathbf{D}$ with $|X|=K$. The dependence on $d$ in the constant is tight for such small sampling sets, which arise naturally for example when studying polynomials of bounded degree coming from functions on products of cyclic groups. As an application we obtain a proof of the cyclic group Bohnenblust-Hille inequality with an explicit constant $O(\log K)^{2d}$.

2310.08111

Junlong Chen

Junlong Chen, Zhaoyang Qiu, Yanbin Tang

Homogenization of the distribution-dependent stochastic abstract fluid models

We corrected some minor errors

math.AP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this paper, we study the homogenization of the distribution-dependent stochastic abstract fluid models by combining the $two\!-\!scale$ convergence and martingale representative approach. A general framework of the homogenization research is established for stochastic abstract fluid models which are the type of nonlinear partial differential equations including the (distribution-dependent) stochastic Navier-Stokes equations, stochastic magneto-hydrodynamic equations, stochastic Boussinesq equations, stochastic micropolar equations, stochastic Allen-Cahn equations.

2310.12446

Jieao Zhu Mr.

Jieao Zhu, Zhongzhichao Wan, Linglong Dai, and Tie Jun Cui

Electromagnetic Information Theory-Based Statistical Channel Model for Improved Channel Estimation

Electromagnetic information theory (EIT) is an emerging interdisciplinary subject, aiming at providing a unified analytical framework for wireless systems as well as guiding practical system design. This paper answers the question: "Whether can we improve wireless communication systems via EIT"?

cs.IT eess.SP math.IT

http://creativecommons.org/publicdomain/zero/1.0/

Electromagnetic information theory (EIT) is an emerging interdisciplinary subject that integrates classical Maxwell electromagnetics and Shannon information theory. The goal of EIT is to uncover the information transmission mechanisms from an electromagnetic (EM) perspective in wireless systems. Existing works on EIT are mainly focused on the analysis of EM channel characteristics, degrees-of-freedom, and system capacity. However, these works do not clarify how to integrate EIT knowledge into the design and optimization of wireless systems. To fill in this gap, in this paper, we propose an EIT-based statistical channel model with simplified parameterization. Thanks to the simplified closed-form expression of the EMCF, it can be readily applied to various channel modeling and inference tasks. Specifically, by averaging the solutions of Maxwell's equations over a tunable von Mises distribution, we obtain a spatio-temporal correlation function (STCF) model of the EM channel, which we name as the EMCF. Furthermore, by tuning the parameters of the EMCF, we propose an EIT-based covariance estimator (EIT-Cov) to accurately capture the channel covariance. Since classical MMSE estimators can exploit prior information contained in the channel covariance matrix, we further propose the EIT-MMSE channel estimator by substituting EMCF for the covariance matrix. Simulation results show that both the proposed EIT-Cov covariance estimator and the EIT-MMSE channel estimator outperform their baseline algorithms, thus proving that EIT is beneficial to wireless communication systems.

2310.13008

Haotian Zhou

Haotian Zhou, Tingkai Liu, Qianli Ma, Yufeng Zhang, Jianbo Yuan, Pengfei Liu, Yang You and Hongxia Yang

DavIR: Data Selection via Implicit Reward for Large Language Models

cs.LG cs.AI cs.CL

http://creativecommons.org/licenses/by/4.0/

We introduce DavIR, a model-based data selection method for post-training Large Language Models. DavIR generalizes Reducible Holdout Loss to core-set selection problem of causal language modeling, and quantifies the learnability of a given datum with respect to a pre-trained LLM based on relative reduction in loss during fine-tuning, a metric we show to be closely related to the implicit reward model described in Direct Preference Optimization (DPO). We show that 6% of Alpaca dataset selected with DavIR can steer both the LLaMA and Gemma model family to produce superior performance compared to the same models trained on the full 52K dataset. We also show that Alpaca dataset compressed with DavIR can be combined with GSM8K dataset to effectively balance open-domain freeform QA and mathematical reasoning capabilities. Finally, we apply the DavIR objective to DPO and develop a normalized DavIR-DPO objective which improves alignment performance of Zephyr-7B-SFT model by 8% (relative) on AlpacaEval, compared against training on vanilla DPO objective.

2310.13174

Ce Jin

Timothy M. Chan, Ce Jin, Virginia Vassilevska Williams, Yinzhan Xu

Faster Algorithms for Text-to-Pattern Hamming Distances

Appeared in FOCS 2023. Abstract shortened to fit arXiv requirements. v3: Fixed a mistake in the proof of Lemma 5.3 (and changed the auxiliary Lemma 5.2). v2: added reference and discussion related to Lemma 2.2 and Appendix B

cs.DS

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We study the classic Text-to-Pattern Hamming Distances problem: given a pattern $P$ of length $m$ and a text $T$ of length $n$, both over a polynomial-size alphabet, compute the Hamming distance between $P$ and $T[i\, .\, . \, i+m-1]$ for every shift $i$, under the standard Word-RAM model with $\Theta(\log n)$-bit words. - We provide an $O(n\sqrt{m})$ time Las Vegas randomized algorithm for this problem, beating the decades-old $O(n \sqrt{m \log m})$ running time [Abrahamson, SICOMP 1987]. We also obtain a deterministic algorithm, with a slightly higher $O(n\sqrt{m}(\log m\log\log m)^{1/4})$ running time. Our randomized algorithm extends to the $k$-bounded setting, with running time $O\big(n+\frac{nk}{\sqrt{m}}\big)$, removing all the extra logarithmic factors from earlier algorithms [Gawrychowski and Uzna\'{n}ski, ICALP 2018; Chan, Golan, Kociumaka, Kopelowitz and Porat, STOC 2020]. - For the $(1+\epsilon)$-approximate version of Text-to-Pattern Hamming Distances, we give an $\tilde{O}(\epsilon^{-0.93}n)$ time Monte Carlo randomized algorithm, beating the previous $\tilde{O}(\epsilon^{-1}n)$ running time [Kopelowitz and Porat, FOCS 2015; Kopelowitz and Porat, SOSA 2018]. Our approximation algorithm exploits a connection with $3$SUM, and uses a combination of Fredman's trick, equality matrix product, and random sampling; in particular, we obtain new results on approximate counting versions of $3$SUM and Exact Triangle, which may be of independent interest. Our exact algorithms use a novel combination of hashing, bit-packed FFT, and recursion; in particular, we obtain a faster algorithm for computing the sumset of two integer sets, in the regime when the universe size is close to quadratic in the number of elements. We also prove a fine-grained equivalence between the exact Text-to-Pattern Hamming Distances problem and a range-restricted, counting version of $3$SUM.

2310.14778

Jinzheng Zhao

Jinzheng Zhao, Yong Xu, Xinyuan Qian, Davide Berghi, Peipei Wu, Meng Cui, Jianyuan Sun, Philip J.B. Jackson and Wenwu Wang

Audio-Visual Speaker Tracking: Progress, Challenges, and Future Directions

cs.MM cs.SD eess.AS

http://creativecommons.org/licenses/by/4.0/

Audio-visual speaker tracking has drawn increasing attention over the past few years due to its academic values and wide application. Audio and visual modalities can provide complementary information for localization and tracking. With audio and visual information, the Bayesian-based filter can solve the problem of data association, audio-visual fusion and track management. In this paper, we conduct a comprehensive overview of audio-visual speaker tracking. To our knowledge, this is the first extensive survey over the past five years. We introduce the family of Bayesian filters and summarize the methods for obtaining audio-visual measurements. In addition, the existing trackers and their performance on AV16.3 dataset are summarized. In the past few years, deep learning techniques have thrived, which also boosts the development of audio visual speaker tracking. The influence of deep learning techniques in terms of measurement extraction and state estimation is also discussed. At last, we discuss the connections between audio-visual speaker tracking and other areas such as speech separation and distributed speaker tracking.

2310.16727

Ronald Schnitzer

Ronald Schnitzer, Andreas Hapfelmeier, Sven Gaube, Sonja Zillner

AI Hazard Management: A framework for the systematic management of root causes for AI risks

10.1007/978-981-99-9836-4_27

cs.LG

http://creativecommons.org/licenses/by/4.0/

Recent advancements in the field of Artificial Intelligence (AI) establish the basis to address challenging tasks. However, with the integration of AI, new risks arise. Therefore, to benefit from its advantages, it is essential to adequately handle the risks associated with AI. Existing risk management processes in related fields, such as software systems, need to sufficiently consider the specifics of AI. A key challenge is to systematically and transparently identify and address AI risks' root causes - also called AI hazards. This paper introduces the AI Hazard Management (AIHM) framework, which provides a structured process to systematically identify, assess, and treat AI hazards. The proposed process is conducted in parallel with the development to ensure that any AI hazard is captured at the earliest possible stage of the AI system's life cycle. In addition, to ensure the AI system's auditability, the proposed framework systematically documents evidence that the potential impact of identified AI hazards could be reduced to a tolerable level. The framework builds upon an AI hazard list from a comprehensive state-of-the-art analysis. Also, we provide a taxonomy that supports the optimal treatment of the identified AI hazards. Additionally, we illustrate how the AIHM framework can increase the overall quality of a power grid AI use case by systematically reducing the impact of identified hazards to an acceptable level.

2310.17955

Beno\^it Darqui\'e

D B A Tran (LPL), Olivier Lopez, M Manceau (LPL), A Goncharov (LPL), Michel Abgrall, H Alvarez-Martinez (SYRTE), R Le Targat (SYRTE), E Cantin (SYRTE), P.-E Pottie (SYRTE), A Amy-Klein (LPL), B Darqui\'e (LPL)

Near-to mid-IR spectral purity transfer with a tunable frequency comb: methanol frequency metrology over a record frequency span

14 pages, 5 figures

APL Photonics 9, 030801 (2024)

10.1063/5.0170227

physics.atom-ph physics.ins-det physics.optics

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We report the upgrade and operation of a frequency-comb-assisted high-resolution mid-infrared molecular spectrometer allowing us to combine high spectral purity, SI-traceability, wide tunability and high sensitivity. An optical frequency comb is used to transfer the spectral purity of a SI-traceable 1.54 $\mu$m metrology-grade frequency reference to a 10.3 $\mu$m quantum cascade laser (QCL). The near-infrared reference is operated at the French time/frequency metrology institute, calibrated there to primary frequency standards, and transferred to Laboratoire de Physique des Lasers via the REFIMEVE fiber network. The QCL exhibits a line width of $\delta\nu\sim 0.1$ Hz, a sub-$10^{-15}$ relative frequency stability from 0.1 to 10 s and its frequency is traceable to the SI with a total relative uncertainty better than $4\times10^{-14}$ after 1-s averaging time. We have developed the instrumentation allowing comb modes to be continuously tuned over 9 GHz resulting in a QCL of record spectral purity uninterruptedly tunable at the precision of the reference over an unprecedented span of $\Delta\nu$ = 1.4 GHz. We have used our apparatus to conduct sub-Doppler spectroscopy of methanol in a multi-pass cell, demonstrating state-of-art frequency uncertainties down to the few kilohertz level ($\sim10^{-10}$ in relative value). We have observed weak intensity resonances unreported so far, resolved subtle doublets never seen before and brought to light discrepancies with the HITRAN database. This demonstrates the potential of our apparatus for probing subtle internal molecular processes, building accurate spectroscopic models of polyatomic molecules of atmospheric or astrophysical interest, and carrying out precise spectroscopic tests of fundamental physics.

2310.19371

Yoav Zimhony

Yoav Zimhony

Commutative Control Data for Smoothly Locally Trivial Stratified Spaces

50 pages

Transformation Groups (2024)

10.1007/s00031-024-09893-w

math.DG math.SG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

For a compact Lie group G and a Hamiltonian G-space M with momentum map $\mu:M \to g^*$, we prove that the zero level set $\mu^{-1}(0)$ and the critical set of the norm-squared momentum map are neighbourhood smooth weak deformation retracts. To this end we show that these subsets, stratified by orbit types, satisfy a condition stronger than Whitney (B) regularity - smooth local triviality with conical fibers. Using this condition we construct control data in the sense of Mather with the additional properties that the fiber-wise multiplications by scalars, coming from the tubular neighbourhood structures, preserve strata and commute with each other. We use that control data to obtain the neighbourhood smooth weak deformation retraction. Finally, such structures for the zero level set $\mu^{-1}(0)$ reduce to similar structures for the reduced space $\mu^{-1}(0)/G$, yielding a similar result for the reduced space and its stratified subspaces.

2310.20129

Norhan Eassa

Norhan M. Eassa, Mahmoud M. Moustafa, Arnab Banerjee, Jeffrey Cohn

Gibbs state sampling via cluster expansions

8 pages, 8 figures, and supplementary material

10.1038/s41534-024-00887-w

quant-ph cond-mat.other

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Gibbs states (i.e., thermal states) can be used for several applications such as quantum simulation, quantum machine learning, quantum optimization, and the study of open quantum systems. Moreover, semi-definite programming, combinatorial optimization problems, and training quantum Boltzmann machines can all be addressed by sampling from well-prepared Gibbs states. With that, however, comes the fact that preparing and sampling from Gibbs states on a quantum computer are notoriously difficult tasks. Such tasks can require large overhead in resources and/or calibration even in the simplest of cases, as well as the fact that the implementation might be limited to only a specific set of systems. We propose a method based on sampling from a quasi-distribution consisting of tensor products of mixed states on local clusters, i.e., expanding the full Gibbs state into a sum of products of local "Gibbs-cumulant" type states easier to implement and sample from on quantum hardware. We begin with presenting results for 4-spin linear chains with XY spin interactions, for which we obtain the $ZZ$ dynamical spin-spin correlation functions. We also present the results of measuring the specific heat of the 8-spin chain Gibbs state $\rho_8$.

2311.02786

Travis Waller

S. Travis Waller, Amalia Polydoropoulou, Leandros Tassiulas, Athanasios Ziliaskopoulos, Sisi Jian, Susann Wagenknecht, Georg Hirte, Satish Ukkusuri, Gitakrishnan Ramadurai, Tomasz Bednarz

Mobility as a Resource (MaaR) for resilient human-centric automation: a vision paper

Data Science for Transportation 7, 1 (2025)

10.1007/s42421-024-00115-z

eess.SY cs.SY

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

With technological advances, mobility has been moving from a product (i.e., traditional modes and vehicles), to a service (i.e., Mobility as a Service, MaaS). However, as observed in other fields (e.g. cloud computing resource management) we argue that mobility will evolve from a service to a resource (i.e., Mobility as a Resource, MaaR). Further, due to increasing scarcity of shared mobility spaces across traditional and emerging modes, the transition must be viewed within the critical need for ethical and equitable solutions for the traveling public (i.e., research is needed to avoid hyper-market driven outcomes for society). The evolution of mobility into a resource requires novel conceptual frameworks, technologies, processes and perspectives of analysis. A key component of the future MaaR system is the technological capacity to observe, allocate and manage (in real-time) the smallest envisionable units of mobility (i.e., atomic units of mobility capacity) while providing prioritized attention to human movement and ethical metrics related to access, consumption and impact. To facilitate research into the envisioned future system, this paper proposes initial frameworks which synthesize and advance methodologies relating to highly dynamic capacity reservation systems. Future research requires synthesis across transport network management, demand behavior, mixed-mode usage, and equitable mobility.

2311.04686

Zhenyu Liao

Zhanbo Feng, Yuanjie Wang, Jie Li, Fan Yang, Jiong Lou, Tiebin Mi, Robert. C. Qiu, Zhenyu Liao

Robust and Communication-Efficient Federated Domain Adaptation via Random Features

22 pages, 7 figures, 17 tables, accepted by IEEE Trans. KDE

10.1109/TKDE.2024.3510296

cs.LG cs.DC stat.ML

http://creativecommons.org/licenses/by/4.0/

Modern machine learning (ML) models have grown to a scale where training them on a single machine becomes impractical. As a result, there is a growing trend to leverage federated learning (FL) techniques to train large ML models in a distributed and collaborative manner. These models, however, when deployed on new devices, might struggle to generalize well due to domain shifts. In this context, federated domain adaptation (FDA) emerges as a powerful approach to address this challenge. Most existing FDA approaches typically focus on aligning the distributions between source and target domains by minimizing their (e.g., MMD) distance. Such strategies, however, inevitably introduce high communication overheads and can be highly sensitive to network reliability. In this paper, we introduce RF-TCA, an enhancement to the standard Transfer Component Analysis approach that significantly accelerates computation without compromising theoretical and empirical performance. Leveraging the computational advantage of RF-TCA, we further extend it to FDA setting with FedRF-TCA. The proposed FedRF-TCA protocol boasts communication complexity that is independent of the sample size, while maintaining performance that is either comparable to or even surpasses state-of-the-art FDA methods. We present extensive experiments to showcase the superior performance and robustness (to network condition) of FedRF-TCA.

2311.05444

Maximilian Kaipel

Maximilian Kaipel

The category of a partitioned fan

31 pages, comments welcome. v2: Sec. 4.4 and Thm. 6.20 added, Prop. 4.14, Thm. 6.21 removed, v3: journal accepted manuscript with major generalisations of the main theorems concerning faithful functors (Thm. 1.4 and Thm. 1.5)

math.RT

http://creativecommons.org/licenses/by/4.0/

In this paper, we introduce the notion of an admissible partition of a simplicial polyhedral fan and define the category of a partitioned fan as a generalisation of the $\tau$-cluster morphism category of a finite-dimensional algebra. This establishes a complete lattice of categories around the $\tau$-cluster morphism category, which is closely tied to the fan structure. We prove that the classifying spaces of these categories are cube complexes, which reduces the process of determining if they are $K(\pi,1)$ spaces to three sufficient conditions. We characterise when these conditions are satisfied for fans in $\mathbb{R}^2$ and prove that the first one, the existence of a certain faithful functor, is satisfied for hyperplane arrangements whose normal vectors lie in the positive orthant. As a consequence we obtain a new infinite class of algebras for which the $\tau$-cluster morphism category admits a faithful functor and for which the cube complexes are $K(\pi,1)$ spaces. In the final section we also offer a new algebraic proof of the relationship between an algebra and its $g$-vector fan.

2311.05999

Roberto Ognibene

Veronica Felli, Lorenzo Liverani, Roberto Ognibene

Quantitative spectral stability for the Neumann Laplacian in domains with small holes

math.AP math.SP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The aim of the present paper is to investigate the behavior of the spectrum of the Neumann Laplacian in domains with little holes excised from the interior. More precisely, we consider the eigenvalues of the Laplacian with homogeneous Neumann boundary conditions on a bounded, Lipschitz domain. Then, we singularly perturb the domain by removing Lipschitz sets which are "small" in a suitable sense and satisfy a uniform extension property. In this context, we provide an asymptotic expansion for all the eigenvalues of the perturbed problem which are converging to simple eigenvalues of the limit one. The eigenvalue variation turns out to depend on a geometric quantity resembling the notion of (boundary) torsional rigidity: understanding this fact is one of the main contributions of the present paper. In the particular case of a hole shrinking to a point, through a fine blow-up analysis, we identify the exact vanishing order of such a quantity and we establish some connections between the location of the hole and the sign of the eigenvalue variation.

2311.06069

Paul Mycek

J\'er\'emy Briant (1 and 2), Paul Mycek (2), Mayeul Destouches (3), Olivier Goux (2), Serge Gratton (1 and 4), Selime G\"urol (2), Ehouarn Simon (1), Anthony T. Weaver (2) ((1) Irit CNRS-INP, (2) CECI CNRS-Cerfacs, (3) Met Office, (4) ANITI)

A filtered multilevel Monte Carlo method for estimating the expectation of cell-centered discretized random fields

math.NA cs.NA

http://creativecommons.org/licenses/by/4.0/

We investigate the use of multilevel Monte Carlo (MLMC) methods for estimating the expectation of discretized random fields. Specifically, we consider a setting in which the input and output vectors of numerical simulators have inconsistent dimensions across the multilevel hierarchy. This requires the introduction of grid transfer operators borrowed from multigrid methods. By adapting mathematical tools from multigrid methods, we perform a theoretical spectral analysis of the MLMC estimator of the expectation of discretized random fields, in the specific case of linear, symmetric and circulant simulators. We then propose filtered MLMC (F-MLMC) estimators based on a filtering mechanism similar to the smoothing process of multigrid methods, and we show that the filtering operators improve the estimation of both the small- and large-scale components of the variance, resulting in a reduction of the total variance of the estimator. Next, the conclusions of the spectral analysis are experimentally verified with a one-dimensional illustration. Finally, the proposed F-MLMC estimator is applied to the problem of estimating the discretized variance field of a diffusion-based covariance operator, which amounts to estimating the expectation of a discretized random field. The numerical experiments support the conclusions of the theoretical analysis even with non-linear simulators, and demonstrate the improvements brought by the F-MLMC estimator compared to both a crude MC and an unfiltered MLMC estimator.

2311.07326

Yanjie Li

Yanjie Li, Weijun Li, Lina Yu, Min Wu, Jinyi Liu, Wenqiang Li, Meilan Hao, Shu Wei, Yusong Deng

MetaSymNet: A Tree-like Symbol Network with Adaptive Architecture and Activation Functions

This work has been accepted by AAAI2025

cs.LG cs.AI

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Mathematical formulas serve as the means of communication between humans and nature, encapsulating the operational laws governing natural phenomena. The concise formulation of these laws is a crucial objective in scientific research and an important challenge for artificial intelligence (AI). While traditional artificial neural networks (MLP) excel at data fitting, they often yield uninterpretable black box results that hinder our understanding of the relationship between variables x and predicted values y. Moreover, the fixed network architecture in MLP often gives rise to redundancy in both network structure and parameters. To address these issues, we propose MetaSymNet, a novel neural network that dynamically adjusts its structure in real-time, allowing for both expansion and contraction. This adaptive network employs the PANGU meta function as its activation function, which is a unique type capable of evolving into various basic functions during training to compose mathematical formulas tailored to specific needs. We then evolve the neural network into a concise, interpretable mathematical expression. To evaluate MetaSymNet's performance, we compare it with four state-of-the-art symbolic regression algorithms across more than 10 public datasets comprising 222 formulas. Our experimental results demonstrate that our algorithm outperforms others consistently regardless of noise presence or absence. Furthermore, we assess MetaSymNet against MLP and SVM regarding their fitting ability and extrapolation capability, these are two essential aspects of machine learning algorithms. The findings reveal that our algorithm excels in both areas. Finally, we compared MetaSymNet with MLP using iterative pruning in network structure complexity. The results show that MetaSymNet's network structure complexity is obviously less than MLP under the same goodness of fit.

2311.07419

Philipp Sterzinger

Philipp Sterzinger and Ioannis Kosmidis

Diaconis-Ylvisaker prior penalized likelihood for $p/n \to \kappa \in (0,1)$ logistic regression

25 pages, 8 figures, pdf attached

math.ST stat.TH

http://creativecommons.org/licenses/by/4.0/

We characterise the behaviour of the maximum Diaconis-Ylvisaker prior penalized likelihood estimator in high-dimensional logistic regression, where the number of covariates is a fraction $\kappa \in (0,1)$ of the number of observations $n$, as $n \to \infty$. We derive the estimator's aggregate asymptotic behaviour under this proportional asymptotic regime, when covariates are independent normal random variables with mean zero and the linear predictor has asymptotic variance $\gamma^2$. From this foundation, we devise adjusted $Z$-statistics, penalized likelihood ratio statistics, and aggregate asymptotic results with arbitrary covariate covariance. While the maximum likelihood estimate asymptotically exists only for a narrow range of $(\kappa, \gamma)$ values, the maximum Diaconis-Ylvisaker prior penalized likelihood estimate not only exists always but is also directly computable using maximum likelihood routines. Thus, our asymptotic results also hold for $(\kappa, \gamma)$ values where results for maximum likelihood are not attainable, with no overhead in implementation or computation. We study the estimator's shrinkage properties, compare it to alternative estimation methods that can operate with proportional asymptotics, and present procedures for the estimation of unknown constants that describe the asymptotic behaviour of our estimator. We also provide a conjecture about the behaviour of our estimator when an intercept parameter is present in the model. We present results from extensive numerical studies to demonstrate the theoretical advances and strong evidence to support the conjecture, and illustrate the methodology we put forward through the analysis of a real-world data set on digit recognition.

2311.08023

David Bevan

David Bevan, Gi-Sang Cheon and Sergey Kitaev

On naturally labelled posets and permutations avoiding 12-34

22 pages; to appear in the European Journal of Combinatorics

math.CO

http://creativecommons.org/licenses/by/4.0/

A partial order $\prec$ on $[n]$ is naturally labelled (NL) if $x\prec y$ implies $x<y$. We establish a bijection between {3, 2+2}-free NL posets and 12-34-avoiding permutations, determine functional equations satisfied by their generating function, and use series analysis to investigate their asymptotic growth, presenting evidence of stretched exponential behaviour. We also exhibit bijections between 3-free NL posets and various other objects, and determine their generating function. The connection between our results and a hierarchy of combinatorial objects related to interval orders is described.

2311.09459

Jacopo Castellini Ph.D.

Rafael F. Cunha, Jacopo Castellini, Johan Peralez, Jilles S. Dibangoye

On Convex Optimal Value Functions For POSGs

Currently under review at JAIR

cs.MA cs.LG

http://creativecommons.org/licenses/by/4.0/

Multi-agent planning and reinforcement learning can be challenging when agents cannot see the state of the world or communicate with each other due to communication costs, latency, or noise. Partially Observable Stochastic Games (POSGs) provide a mathematical framework for modelling such scenarios. This paper aims to improve the efficiency of planning and reinforcement learning algorithms for POSGs by identifying the underlying structure of optimal state-value functions. The approach involves reformulating the original game from the perspective of a trusted third party who plans on behalf of the agents simultaneously. From this viewpoint, the original POSGs can be viewed as Markov games where states are occupancy states, \ie posterior probability distributions over the hidden states of the world and the stream of actions and observations that agents have experienced so far. This study mainly proves that the optimal state-value function is a convex function of occupancy states expressed on an appropriate basis in all zero-sum, common-payoff, and Stackelberg POSGs.

2311.10029

Maksym Tsizh

Vitalii Tymchyshyn, Maksym Tsizh, Franco Vazza, Marco Baldi

On the evolution of Betti curves in the Cosmic web

21 pages, 8 figures

Journal of Physical Studies 28(4), Article 4903 (2024)

10.30970/jps.28.4903

astro-ph.CO

http://creativecommons.org/licenses/by-sa/4.0/

In this work, we study the evolution of Betti curves obtained by persistent-homological analysis of point clouds formed by halos in different cosmological $N$-body simulations. We show that they can be approximated with a scaled log-normal distribution function with reasonable precision. Our analysis shows that the shapes and maximums of Betti curves exhibit dependence on the mass range of the selected subpopulation of halos. Still, at the same time, the resolution of a simulation does not play any significant role, provided that the mass distribution of simulated halos is complete down to a given mass scale. Besides, we study how Betti curves change with the evolution of the Universe, i.e., their dependence on redshift. Sampling subpopulations of halos within certain mass ranges up to redshift $z=2.5$ yields a surprisingly small difference between corresponding Betti curves. We propose that this may be an indicator of the existence of a new specific topological invariant in the structure of the Universe.

2311.13638

Christopher Cain

Christopher Cain, Anson D'Aloisio, Garett Lopez, Nakul Gangolli, Joshua T. Roth

On the rise and fall of galactic ionizing output at the end of reionization

20 pages, 15 figures, accepted for publication in MNRAS

astro-ph.CO astro-ph.GA

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Quasar absorption spectra measurements suggest that reionization proceeded rapidly, ended late at $z \sim 5.5$, and was followed by a flat ionizing background evolution. Simulations that reproduce this behavior often rely on a fine tuned galaxy ionizing emissivity, which peaks at $z \sim 6-7$ and drops a factor of $1.5-2.5$ by $z \sim 5$. This is puzzling since the abundance of galaxies is observed to grow monotonically during this period. Explanations for this include effects such as dust obscuration of ionizing photon escape and feedback from photo-heating of the IGM. We explore the possibility that this drop in emissivity is instead an artifact of one or more modeling deficiencies in reionization simulations. These include possibly incorrect assumptions about the ionizing spectrum and/or inaccurate modeling of IGM clumping. Our results suggest that the need for a drop could be alleviated if simulations are underestimating the IGM opacity from massive, star-forming halos. Other potential modeling issues either have a small effect or require a steeper drop when remedied. We construct an illustrative model in which the emissivity is nearly flat at reionization's end, evolving only $\sim 0.05$ dex at $5 < z < 7$. More realistic scenarios, however, require a $\sim 0.1-0.3$ dex drop. We also study the evolution of the Ly$\alpha$ effective optical depth distribution and compare to recent measurements. We find models that feature a hard ionizing spectrum and/or are driven by faint, low-bias sources most easily reproduce the mean transmission and optical depth distribution of the forest simultaneously.

2311.13932

Raffaele Vitolo

P. Lorenzoni, R. Vitolo

Bi-Hamiltonian structures of KdV type, cyclic Frobenius algebrae and Monge metrics

24 pages

J. Phys. A: Math. Theor. 57 485202 (2024)

10.1088/1751-8121/ad8fe6

math-ph math.MP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We study algebraic and projective geometric properties of Hamiltonian trios determined by a constant coefficient second-order operator and two first-order localizable operators of Ferapontov type. We show that first-order operators are determined by Monge metrics, and define a structure of cyclic Frobenius algebra. Examples include the AKNS system, a $2$-component generalization of Camassa-Holm equation and the Kaup--Broer system. In dimension $2$ the trio is completely determined by two conics of rank at least $2$. We provide a partial classification in dimension $4$.

2311.13959

Yue Song

Yue Song, Wei Wang, Nicu Sebe

RankFeat&RankWeight: Rank-1 Feature/Weight Removal for Out-of-distribution Detection

accepted by T-PAMI. arXiv admin note: substantial text overlap with arXiv:2209.08590

cs.LG cs.CV

http://creativecommons.org/licenses/by-nc-nd/4.0/

The task of out-of-distribution (OOD) detection is crucial for deploying machine learning models in real-world settings. In this paper, we observe that the singular value distributions of the in-distribution (ID) and OOD features are quite different: the OOD feature matrix tends to have a larger dominant singular value than the ID feature, and the class predictions of OOD samples are largely determined by it. This observation motivates us to propose \texttt{RankFeat}, a simple yet effective \emph{post hoc} approach for OOD detection by removing the rank-1 matrix composed of the largest singular value and the associated singular vectors from the high-level feature. \texttt{RankFeat} achieves \emph{state-of-the-art} performance and reduces the average false positive rate (FPR95) by 17.90\% compared with the previous best method. The success of \texttt{RankFeat} motivates us to investigate whether a similar phenomenon would exist in the parameter matrices of neural networks. We thus propose \texttt{RankWeight} which removes the rank-1 weight from the parameter matrices of a single deep layer. Our \texttt{RankWeight}is also \emph{post hoc} and only requires computing the rank-1 matrix once. As a standalone approach, \texttt{RankWeight} has very competitive performance against other methods across various backbones. Moreover, \texttt{RankWeight} enjoys flexible compatibility with a wide range of OOD detection methods. The combination of \texttt{RankWeight} and \texttt{RankFeat} refreshes the new \emph{state-of-the-art} performance, achieving the FPR95 as low as 16.13\% on the ImageNet-1k benchmark. Extensive ablation studies and comprehensive theoretical analyses are presented to support the empirical results. Code is publicly available via \url{https://github.com/KingJamesSong/RankFeat}.

2311.14503

Daniel Margineda

Daniel Margineda and Alessandro Crippa and Elia Strambini and Laura Borgongino and Alessandro Paghi and Giorgio de Simoni and Lucia Sorba andYuri Fukaya and Maria Teresa Mercaldo and Carmine Ortix and Mario Cuoco and Francesco Giazotto

Back-action supercurrent diodes

cond-mat.mes-hall cond-mat.supr-con

http://creativecommons.org/licenses/by/4.0/

Back-action refers to a response that retro-acts on a system to tailor its properties with respect to an external stimulus. This effect is at the heart of many electronic devices such as amplifiers, oscillators, and sensors. Here, we demonstrate that back-action can be exploited to achieve non-reciprocal transport in superconducting circuits. In our devices, dissipationless current flows in one direction whereas dissipative transport occurs in the opposite direction. Supercurrent diodes presented so far rely on magnetic elements or vortices to mediate charge transport or external magnetic fields to break time-reversal symmetry. Back-action solely turns a conventional reciprocal superconducting weak link with no asymmetry between the current bias directions into a rectifier, where the critical current amplitude depends on the bias sign. The self-interaction of the supercurrent stems from the gate tunability of the critical current in metallic and semiconducting systems, which promotes nearly ideal magnetic field-free rectification with selectable polarity.

2311.14555

Amina Mecherbet

Matthieu Bonnivard and Amina Mecherbet

On the instability of travelling wave solutions for the transport-Stokes equation

Mathematical and numerical investigation of the instability of the linearized system have been completed in sections 3.3 and 3.4 since the last update

10.4310/CMS.241217211132

math.AP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this paper, we investigate the instability of the spherical travelling wave solutions for the Transport-Stokes system in $\mathbb{R}^3$. First, a classical scaling argument ensures instability among all probability measures for the Wasserstein metric and the $L^1$ norm. Secondly, we address the instability among patch solutions with a perturbed surface. To this end, we study the linearized system of a contour dynamics equation derived in [18] in the case where the support of the patch is axisymmetric and described by spherical parametrization. We investigate numerically the existence of positive eigenvalues, which ensures the instability of the linearized system. Eventually we recover numerically the instability of the travelling wave by solving the Transport-Stokes equation using a finite element method on FreeFem.

2311.15224

Long Huang

Long Huang, Yuanshou Cao, Dachun Yang and Ciqiang Zhuo

Sharp Poincar\'e--Sobolev Inequalities of Choquet--Lorentz Integrals with Respect to Hausdorff Contents on Bounded John Domains

23 pages

math.FA math.AP math.CA

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Let $\Omega$ be a bounded John domain in $\mathbb R^n$ with $n\ge 2$, and let $\mathcal{H}_{\infty }^{\delta}$ denote the Hausdorff content of dimension $\delta\in (0,n]$. In this article, the authors prove the Poincar\'e and the Poincar\'e--Sobolev inequalities, with sharp ranges of indices, on Choquet--Lorentz integrals with respect to $\mathcal{H}_{\infty }^{\delta}$ for all continuously differentiable functions on $\Omega$. These results not only extend the recent Poincar\'e and Poincar\'e--Sobolev inequalities to the Choquet--Lorentz integrals, but also provide some endpoint estimates (weak type) in the critical case. One of the main novelties exists in that, to achieve the goals, the authors develop some new tools associated with Choquet--Lorentz integrals on $\mathcal{H}_{\infty }^{\delta}$, such as the fractional Hardy--Littlewood maximal inequality and the Hedberg-type pointwise estimate on the Riesz potential. As an application, the authors obtain the sharp boundedness of the Riesz potential on Choquet--Lorentz integrals. Moreover, even for classical Lorentz integrals, these Poincar\'e and Poincar\'e--Sobolev inequalities are also new.

2311.16208

He Cao

He Cao, Zijing Liu, Xingyu Lu, Yuan Yao, Yu Li

InstructMol: Multi-Modal Integration for Building a Versatile and Reliable Molecular Assistant in Drug Discovery

q-bio.BM cs.AI cs.LG

http://creativecommons.org/licenses/by/4.0/

The rapid evolution of artificial intelligence in drug discovery encounters challenges with generalization and extensive training, yet Large Language Models (LLMs) offer promise in reshaping interactions with complex molecular data. Our novel contribution, InstructMol, a multi-modal LLM, effectively aligns molecular structures with natural language via an instruction-tuning approach, utilizing a two-stage training strategy that adeptly combines limited domain-specific data with molecular and textual information. InstructMol showcases substantial performance improvements in drug discovery-related molecular tasks, surpassing leading LLMs and significantly reducing the gap with specialized models, thereby establishing a robust foundation for a versatile and dependable drug discovery assistant.

2311.17605

Xingjian Ma

Xingjian Ma and Yang Liu

Improving the Balance of Unobserved Covariates From Information Theory in Multi-Arm Randomization with Unequal Allocation Ratio

The article's structure and theoretical framework have undergone substantial revisions to improve clarity and rigor. Additionally, the numerical experiments have been entirely re-implemented to ensure consistency with the updated theoretical developments. We plan to resubmit the revised version after completing these improvements

stat.AP stat.ME

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Multi-arm randomization has increasingly widespread applications recently and it is also crucial to ensure that the distributions of important observed covariates as well as the potential unobserved covariates are similar and comparable among all the treatment. However, the theoretical properties of unobserved covariates imbalance in multi-arm randomization with unequal allocation ratio remains unknown. In this paper, we give a general framework analysing the moments and distributions of unobserved covariates imbalance and apply them into different procedures including complete randomization (CR), stratified permuted block (STR-PB) and covariate-adaptive randomization (CAR). The general procedures of multi-arm STR-PB and CAR with unequal allocation ratio are also proposed. In addition, we introduce the concept of entropy to measure the correlation between discrete covariates and verify that we could utilize the correlation to select observed covariates to help better balance the unobserved covariates.

2311.18512

Aritra Bhowmik

Aritra Bhowmik, Pascal Mettes, Martin R. Oswald, Cees G. M. Snoek

Union-over-Intersections: Object Detection beyond Winner-Takes-All

17 pages, 6 figures, 12 tables

cs.CV cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper revisits the problem of predicting box locations in object detection architectures. Typically, each box proposal or box query aims to directly maximize the intersection-over-union score with the ground truth, followed by a winner-takes-all non-maximum suppression where only the highest scoring box in each region is retained. We observe that both steps are sub-optimal: the first involves regressing proposals to the entire ground truth, which is a difficult task even with large receptive fields, and the second neglects valuable information from boxes other than the top candidate. Instead of regressing proposals to the whole ground truth, we propose a simpler approach: regress only to the area of intersection between the proposal and the ground truth. This avoids the need for proposals to extrapolate beyond their visual scope, improving localization accuracy. Rather than adopting a winner-takes-all strategy, we take the union over the regressed intersections of all boxes in a region to generate the final box outputs. Our plug-and-play method integrates seamlessly into proposal-based, grid-based, and query-based detection architectures with minimal modifications, consistently improving object localization and instance segmentation. We demonstrate its broad applicability and versatility across various detection and segmentation tasks.

2312.00326

Zhangcheng Qiang

Zhangcheng Qiang, Weiqing Wang, Kerry Taylor

Agent-OM: Leveraging LLM Agents for Ontology Matching

19 pages, 13 figures, 4 tables

cs.AI cs.CL cs.IR

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Ontology matching (OM) enables semantic interoperability between different ontologies and resolves their conceptual heterogeneity by aligning related entities. OM systems currently have two prevailing design paradigms: conventional knowledge-based expert systems and newer machine learning-based predictive systems. While large language models (LLMs) and LLM agents have revolutionised data engineering and have been applied creatively in many domains, their potential for OM remains underexplored. This study introduces a novel agent-powered LLM-based design paradigm for OM systems. With consideration of several specific challenges in leveraging LLM agents for OM, we propose a generic framework, namely Agent-OM (Agent for Ontology Matching), consisting of two Siamese agents for retrieval and matching, with a set of simple OM tools. Our framework is implemented in a proof-of-concept system. Evaluations of three Ontology Alignment Evaluation Initiative (OAEI) tracks over state-of-the-art OM systems show that our system can achieve results very close to the long-standing best performance on simple OM tasks and can significantly improve the performance on complex and few-shot OM tasks.

2312.04725

Simon Rutard

Simon Rutard

Values and derivative values at nonpositive integers of generalized multiple Hurwitz zeta functions, applications to Witten zeta functions

math.NT

http://creativecommons.org/licenses/by/4.0/

We present an analytic continuation of certain multiple zeta functions of generalized Hurwitz type. From this analytic continuation we derive explicit formulas for their values and derivative values at nonpositive integers along a direction. As an application, we provide explicit formulas for some values and derivative values of the Witten zeta functions $\zeta_{\mathfrak{g}_2}$ and $\zeta_{\mathfrak{so}(5)}$. Furthermore, by employing a Meinardus-type theorem, we investigate the asymptotic behavior of the number of $n$-dimensional representations of the exceptional Lie algebra $\mathfrak{g}_2$.

2312.06259

Zhiyi Pan

Zhiyi Pan and Nan Zhang and Wei Gao and Shan Liu and Ge Li

Point Cloud Semantic Segmentation with Sparse and Inhomogeneous Annotations

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Utilizing uniformly distributed sparse annotations, weakly supervised learning alleviates the heavy reliance on fine-grained annotations in point cloud semantic segmentation tasks. However, few works discuss the inhomogeneity of sparse annotations, albeit it is common in real-world scenarios. Therefore, this work introduces the probability density function into the gradient sampling approximation method to qualitatively analyze the impact of annotation sparsity and inhomogeneity under weakly supervised learning. Based on our analysis, we propose an Adaptive Annotation Distribution Network (AADNet) capable of robust learning on arbitrarily distributed sparse annotations. Specifically, we propose a label-aware point cloud downsampling strategy to increase the proportion of annotations involved in the training stage. Furthermore, we design the multiplicative dynamic entropy as the gradient calibration function to mitigate the gradient bias caused by non-uniformly distributed sparse annotations and explicitly reduce the epistemic uncertainty. Without any prior restrictions and additional information, our proposed method achieves comprehensive performance improvements at multiple label rates and different annotation distributions.

2312.08562

Mariusz Tobolski

Piotr M. Hajac, Mariusz Tobolski

The covariant functoriality of graph algebras

17 pages; Definition 3.5 corrected, Example 4.10 added

math.RA

http://creativecommons.org/licenses/by/4.0/

In the standard category of directed graphs, graph morphisms map edges to edges. By allowing graph morphisms to map edges to finite paths (path homomorphisms of graphs), we obtain an ambient category in which we determine subcategories enjoying covariant functors to categories of algebras given by constructions of path algebras, Cohn path algebras, and Leavitt path algebras, respectively. Thus we obtain new tools to unravel homomorphisms between Leavitt path algebras and graph C*-algebras. In particular, a graph-algebraic presentation of the inclusion of the C*-algebra of a quantum real projective plane into the Toeplitz algebra allows us to determine a quantum CW-complex structure of the former. It comes as a mixed-pullback theorem where two $*$-homomorphisms are covariantly induced from path homomorphisms of graphs and the remaining two are contravariantly induced by admissible inclusions of graphs. As a main result and an application of new covariant-induction tools, we prove such a mixed-pullback theorem for arbitrary graphs whose all vertex-simple loops have exits, which substantially enlarges the scope of examples coming from noncommutative topology.

2312.09789

Antonio M. Sudoso

Veronica Piccialli, Jan Schwiddessen, Antonio M. Sudoso

Optimization meets Machine Learning: An Exact Algorithm for Semi-Supervised Support Vector Machines

Mathematical Programming, 2024

10.1007/s10107-024-02175-z

math.OC cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Support vector machines (SVMs) are well-studied supervised learning models for binary classification. In many applications, large amounts of samples can be cheaply and easily obtained. What is often a costly and error-prone process is to manually label these instances. Semi-supervised support vector machines (S3VMs) extend the well-known SVM classifiers to the semi-supervised approach, aiming at maximizing the margin between samples in the presence of unlabeled data. By leveraging both labeled and unlabeled data, S3VMs attempt to achieve better accuracy and robustness compared to traditional SVMs. Unfortunately, the resulting optimization problem is non-convex and hence difficult to solve exactly. In this paper, we present a new branch-and-cut approach for S3VMs using semidefinite programming (SDP) relaxations. We apply optimality-based bound tightening to bound the feasible set. Box constraints allow us to include valid inequalities, strengthening the lower bound. The resulting SDP relaxation provides bounds significantly stronger than the ones available in the literature. For the upper bound, instead, we define a local search exploiting the solution of the SDP relaxation. Computational results highlight the efficiency of the algorithm, showing its capability to solve instances with a number of data points 10 times larger than the ones solved in the literature.

2312.10660

Xufeng Zhang

Jing Xu, Connor Horn, Yu Jiang, Amin Pishehvar, Xinhao Li, Daniel Rosenmann, Xu Han, Miguel Levy, Supratik Guha, Xufeng Zhang

Cryogenic hybrid magnonic circuits based on spalled YIG thin films

6 pages, 4 figures

cond-mat.mes-hall quant-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Yttrium iron garnet (YIG) magnonics has garnered significant research interest because of the unique properties of magnons (quasiparticles of collective spin excitation) for signal processing. In particular, hybrid systems based on YIG magnonics show great promise for quantum information science due to their broad frequency tunability and strong compatibility with other platforms. However, their broad applications have been severely constrained by substantial microwave loss in the gadolinium gallium garnet (GGG) substrate at cryogenic temperatures. In this study, we demonstrate that YIG thin films can be spalled from YIG/GGG samples. Our approach is validated by measuring hybrid devices comprising superconducting resonators and spalled YIG films, which exhibits anti-crossing features that indicate strong coupling between magnons and microwave photons. Such new capability of separating YIG thin films from GGG substrates via spalling, and the integrated superconductor-YIG devices represent a significant advancement for integrated magnonic devices, paving the way for advanced magnon-based coherent information processing.

2312.11973

Haeyong Kang

Haeyong Kang, Jaehong Yoon, Sung Ju Hwang, and Chang D. Yoo

Continual Learning: Forget-free Winning Subnetworks for Video Representations

IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI)

10.1109/TPAMI.2024.3518588

cs.CV cs.AI cs.LG

http://creativecommons.org/licenses/by/4.0/

Inspired by the Lottery Ticket Hypothesis (LTH), which highlights the existence of efficient subnetworks within larger, dense networks, a high-performing Winning Subnetwork (WSN) in terms of task performance under appropriate sparsity conditions is considered for various continual learning tasks. It leverages pre-existing weights from dense networks to achieve efficient learning in Task Incremental Learning (TIL) and Task-agnostic Incremental Learning (TaIL) scenarios. In Few-Shot Class Incremental Learning (FSCIL), a variation of WSN referred to as the Soft subnetwork (SoftNet) is designed to prevent overfitting when the data samples are scarce. Furthermore, the sparse reuse of WSN weights is considered for Video Incremental Learning (VIL). The use of Fourier Subneural Operator (FSO) within WSN is considered. It enables compact encoding of videos and identifies reusable subnetworks across varying bandwidths. We have integrated FSO into different architectural frameworks for continual learning, including VIL, TIL, and FSCIL. Our comprehensive experiments demonstrate FSO's effectiveness, significantly improving task performance at various convolutional representational levels. Specifically, FSO enhances higher-layer performance in TIL and FSCIL and lower-layer performance in VIL.

2312.12377

N Furey

N. Furey

An Algebraic Roadmap of Particle Theories, Part I: General construction

15 pages, 6 figures, First paper in a series; proposed connection to Hodge duality and Jordan algebras via volume elements, updated references

HU-EP-23/64

hep-ph hep-th

http://creativecommons.org/licenses/by/4.0/

Expanding the results of [1], [2], [3], we demonstrate a network of algebraic connections between six well-known particle theories. These are the Spin(10) model, the Georgi-Glashow model, the Pati-Salam model, the Left-Right Symmetric model, the Standard Model both pre- and post-Higgs mechanism. A new inclusion of a quaternionic reflection within the network further differentiates $W^{\pm}$ bosons from the $Z^0$ boson in comparison to the Standard Model. It may introduce subtle new considerations for the phenomenology of electroweak symmetry breaking.

2312.12538

Sae Koyama

Sae Koyama

Constructions of superabundant tropical curves in higher genus

18 pages, 14 figures. v3: contains minor corrections and clarifications. Accepted version, to appear in Bulletin of the LMS

10.1112/blms.13209

math.AG

http://creativecommons.org/licenses/by/4.0/

We construct qualitatively new examples of superabundant tropical curves which are non-realizable in genus $3$ and $4$. These curves are in $\mathbb{R}^3$ and $\mathbb{R}^4$ respectively, and have properties resembling canonical embeddings of genus $3$ and $4$ algebraic curves. In particular, the genus $3$ example is a degree $4$ planar tropical curve, and the genus $4$ example is contained in the product of a tropical line and a tropical conic. They have excess dimension of deformation space equal to $1$. Non-realizability follows by combining this with a dimension calculation for the corresponding space of logarithmic curves.

2312.12965

Jef Laga

Jef Laga and Ari Shnidman

Ceresa cycles of bielliptic Picard curves

accepted version

math.AG math.NT

http://creativecommons.org/licenses/by/4.0/

We show that the Ceresa cycle $\kappa(C_t)$ of the genus $3$ curve $C_t \colon y^3 = x^4 + 2tx^2 + 1$ is torsion if and only if $Q_t=( \sqrt[3]{t^2 -1},t)$ is a torsion point on the elliptic curve $y^2 = x^3 + 1$. This shows that there are infinitely many smooth plane quartic curves over $\mathbb{C}$ (resp. $\mathbb{Q}$) with torsion (resp. infinite order) Ceresa cycle. Over $\overline{\mathbb{Q}}$, we show that the Beilinson--Bloch height of $\kappa(C_t)$ is proportional to the Neron--Tate height of $Q_t$. Thus, the height of $\kappa(C_t)$ is nondegenerate and satisfies a Northcott property. To prove all this, we show that the Chow motive that controls $\kappa(C_t)$ is isomorphic to $\mathfrak{h}^1$ of an appropriate elliptic curve.

2312.13262

Jeff Achter

Jeff Achter and Sebastian Casalaina-Martin and Jonathan Wise

Images of abelian schemes

Complete rewrite with additional coauthor. Patrick Brosnan alerted us to a serious flaw in the first version. New version gives conditions on the base when the image of an abelian scheme is, or is not, necessarily an abelian scheme

math.AG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We provide some conditions for the image of a morphism of abelian schemes to again be an abelian scheme. For context, in characteristic 0, the image is always an abelian scheme; in mixed and positive characteristic the image can fail to be an abelian scheme, and so it is in this setting that the conditions we provide are pertinent.

2312.13263

Jeff Achter

Jeff Achter and Sebastian Casalaina-Martin

Putting the p back in Prym

Major rewrite triggered by issues with V1 of arXiv:2312.13262. Statements for covers of curves are unchanged; some general statements (on norms and complements) now proved under additional hypothesis

math.AG math.NT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

After Jacobians of curves, Prym varieties are perhaps the next most studied abelian varieties. They turn out to be quite useful in a number of contexts. For technical reasons, there does not appear to be any systematic treatment of Prym varieties in characteristic 2, and due to our recent interest in this topic, the purpose of this paper is to fill in that gap. Our main result is a classification of branched covers of curves in characteristic 2 that give rise to Prym varieties. We are also interested in the case of Prym varieties in the relative setting, and so we develop that theory here as well, including an extension of Welters' Criterion.

2312.13926

Yan-Lung Leon Li

Siu-Cheong Lau, Nai-Chung Conan Leung, Yan-Lung Leon Li

Equivariant Lagrangian correspondence and a conjecture of Teleman

Exposition improved

math.SG math.AG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this paper, we study the Floer theory of equivariant Lagrangian correspondences and apply it to deduce a conjecture of Teleman, which finds the relation between the disc potential of an invariant Lagrangian submanifold and that of its quotient. A main step is to extend Fukaya's construction of an $A_\infty$ tri-module for Lagrangian correspondences to Borel spaces. We find that the equivariant obstruction of a Lagrangian correspondence plays an essential role, which leads to quantum corrections in the disc potentials of quotients. We solve the obstruction in the toric setup and find the relation with mirror maps for compact semi-Fano toric manifolds.

2312.15379

Justus Fasse

Justus Fasse and Bart Jacobs

A flexible specification approach for verifying total correctness of fine-grained concurrent modules

cs.PL cs.LO

http://creativecommons.org/licenses/by/4.0/

A well-established approach to proving progress properties such as deadlock-freedom and termination is to associate obligations with threads. For example, in most existing work the proof rule for lock acquisition prescribes a standard usage protocol by burdening the acquiring thread with an obligation to release the lock. The fact that the obligation creation is hardcoded into the acquire operation, however, rules out non-standard clients e.g. where the release happens in a different thread. We overcome this limitation by instead having the blocking operations take the obligation creation operations required for the specific client scenario as arguments. We dub this simple instance of higher-order programming with auxiliary code Sassy. To illustrate Sassy, we extend HeapLang, a simple, higher-order, concurrent programming language with erasable code and state. The resulting language gets stuck if no progress is made. Consequently, we can apply standard safety separation logic to compositionally reason about termination in a fine-grained concurrent setting. We validated Sassy by developing (non-foundational) machine-checked proofs of representative locks -- an unfair Spinlock (competitive succession), a fair Ticketlock (direct handoff succession) and the hierarchically constructed Cohortlock that is starvation-free if the underlying locks are starvation-free -- against our specifications using an encoding of the approach in the VeriFast program verifier for C and Java.

2312.15638

Masako Kishida

Masako Kishida

Risk-Aware Control of Discrete-Time Stochastic Systems: Integrating Kalman Filter and Worst-case CVaR in Control Barrier Functions

This has been presented at IEEE Conference on Decision and Control 2024, pp. 2019-2024. Minor typos in equations (12) - (14) have been fixed

math.OC cs.SY eess.SY

http://creativecommons.org/licenses/by/4.0/

This paper proposes control approaches for discrete-time linear systems subject to stochastic disturbances. It employs Kalman filter to estimate the mean and covariance of the state propagation, and the worst-case conditional value-at-risk (CVaR) to quantify the tail risk using the estimated mean and covariance. The quantified risk is then integrated into a control barrier function (CBF) to derive constraints for controller synthesis, addressing tail risks near safe set boundaries. Two optimization-based control methods are presented using the obtained constraints for half-space and ellipsoidal safe sets, respectively. The effectiveness of the obtained results is demonstrated using numerical simulations.

2312.17702

Luis Manuel

Germ\'an G. Blesio, Luis O. Manuel, Armando A. Aligia

Anisotropy-driven topological quantum phase transition in magnetic impurities

30 pages, 14 figures. Accepted for publication in Materials for Quantum Technology (IOP)

cond-mat.str-el

http://creativecommons.org/licenses/by/4.0/

A few years ago, a topological quantum phase transition (TQPT) has been found in Anderson and Kondo 2-channel spin-1 impurity models that include a hard-axis anisotropy term $DS_z^2$ with $D > 0$. The most remarkable manifestation of the TQPT is a jump in the spectral density of localized electrons, at the Fermi level, from very high to very low values as $D$ is increased. If the two conduction channels are equivalent, the transition takes place at the critical anisotropy $D_c \sim 2.5\; T_K$, where $T_K$ is the Kondo temperature for $D=0$. This jump might be important to develop a molecular transistor. The jump is due to a corresponding one in the Luttinger integral, which has a topological non-trivial value $\pi/2$ for $D > D_c$. Here, we review the main results for the spectral density and highlight the significance of the theory for the interpretation of measurements conducted on magnetic atoms or molecules on metallic surfaces. In these experiments, where $D$ is held constant, the energy scale $T_K$ is manipulated by some parameters. The resulting variation gives rise to a differential conductance $dI/dV$, measured by scanning-tunneling spectroscopy, which is consistent with a TQPT at an intermediate value of $T_K$. We also show that the theory can be extended to integer spin $S>1$ and two-impurity systems. This is also probably true for half-integer spin and non-equivalent channels in some cases.

2401.01500

Sharvaj Kubal

Sharvaj Kubal and Christian Campbell and Elina Robeva

Log-concave Density Estimation with Independent Components

44 pages, 10 figures. Various improvements over the previous version (v1), and substantial reorganization of Section 3. Some missing assumptions required by Theorem 3.10 of the previous version (v1) have now been made explicit (Lemma 3.13 of the current version)

math.ST stat.ME stat.TH

http://creativecommons.org/licenses/by/4.0/

We propose a method for estimating a log-concave density on $\mathbb R^d$ from samples, under the assumption that there exists an orthogonal transformation that makes the components of the random vector independent. While log-concave density estimation is hard both computationally and statistically, the independent components assumption alleviates both issues, while still maintaining a large non-parametric class. We prove that under mild conditions, at most $\tilde{\mathcal{O}}(\epsilon^{-4})$ samples (suppressing constants and log factors) suffice for our proposed estimator to be within $\epsilon$ of the original density in squared Hellinger distance. On the computational front, while the usual log-concave maximum likelihood estimate can be obtained via a finite-dimensional convex program, it is slow to compute -- especially in higher dimensions. We demonstrate through numerical experiments that our estimator can be computed efficiently, making it more practical to use.

2401.04262

Guillermo Corti\~nas

Guillermo Corti\~nas, Roozbeh Hazrat

Classification conjectures for Leavitt path algebras

Final version

Bull. Lond. Math. Soc. (2024) 56:10, 3011-3267

10.1112/blms.13139

math.RA math.KT math.OA

http://creativecommons.org/licenses/by/4.0/

The theory of Leavitt path algebras is intrinsically related, via graphs, to the theory of symbolic dynamics and $C^*$-algebras where the major classification programs have been a domain of intense research in the last 50 years. In this survey article, we gather together current lines of research in the classification of Leavitt path algebras, questions, conjectures, and some of the results about them that have been obtained so far.

2401.04770

Georgios Doulis

Georgios Doulis, Sebastiano Bernuzzi, Wolfgang Tichy

Entropy based flux limiting scheme for conservation laws

13 pages, 15 figures

gr-qc astro-ph.HE

http://creativecommons.org/licenses/by/4.0/

The entropy based flux-limiting (EFL) scheme is a novel approach designed to accurately resolve shocks and discontinuities in special and general relativistic hydrodynamics. By adaptively adjusting the numerical fluxes, the EFL method mitigates oscillations and preserves smooth transition across discontinuities in shock-dominated flows. Here, we extend the applicability of the EFL method beyond special/general relativistic hydrodynamics to scalar conservation laws and show how to treat systems without a thermodynamic entropy. This is an indication that the method has universal applicability to any system of partial differential equations that can be written in conservation form. We also present some further very challenging special/general relativistic hydrodynamics applications of the EFL method.

2401.04979

YongKyung Oh

YongKyung Oh, Dongyoung Lim, Sungil Kim

DualDynamics: Synergizing Implicit and Explicit Methods for Robust Irregular Time Series Analysis

Published at the 39th Annual AAAI Conference on Artificial Intelligence (AAAI 2025)

cs.LG cs.AI

http://creativecommons.org/licenses/by/4.0/

Real-world time series analysis faces significant challenges when dealing with irregular and incomplete data. While Neural Differential Equation (NDE) based methods have shown promise, they struggle with limited expressiveness, scalability issues, and stability concerns. Conversely, Neural Flows offer stability but falter with irregular data. We introduce 'DualDynamics', a novel framework that synergistically combines NDE-based method and Neural Flow-based method. This approach enhances expressive power while balancing computational demands, addressing critical limitations of existing techniques. We demonstrate DualDynamics' effectiveness across diverse tasks: classification of robustness to dataset shift, irregularly-sampled series analysis, interpolation of missing data, and forecasting with partial observations. Our results show consistent outperformance over state-of-the-art methods, indicating DualDynamics' potential to advance irregular time series analysis significantly.

2401.05055

Heng Zhou

Yawen Xiang, Heng Zhou, Chengyang Li, Fangwei Sun, Zhongbo Li and Yongqiang Xie

Deep learning in motion deblurring: current status, benchmarks and future prospects

29 pages, 13 figures, more than 150 papers have been included

10.1007/s00371-024-03632-8

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Motion deblurring is one of the fundamental problems of computer vision and has received continuous attention. The variability in blur, both within and across images, imposes limitations on non-blind deblurring techniques that rely on estimating the blur kernel. As a response, blind motion deblurring has emerged, aiming to restore clear and detailed images without prior knowledge of the blur type, fueled by the advancements in deep learning methodologies. Despite strides in this field, a comprehensive synthesis of recent progress in deep learning-based blind motion deblurring is notably absent. This paper fills that gap by providing an exhaustive overview of the role of deep learning in blind motion deblurring, encompassing datasets, evaluation metrics, and methods developed over the last six years. Specifically, we first introduce the types of motion blur and the fundamental principles of deblurring. Next, we outline the shortcomings of traditional non-blind deblurring algorithms, emphasizing the advantages of employing deep learning techniques for deblurring tasks. Following this, we categorize and summarize existing blind motion deblurring methods based on different backbone networks, including convolutional neural networks, generative adversarial networks, recurrent neural networks, and Transformer networks. Subsequently, we elaborate not only on the fundamental principles of these different categories but also provide a comprehensive summary and comparison of their advantages and limitations. Qualitative and quantitative experimental results conducted on four widely used datasets further compare the performance of SOTA methods. Finally, an analysis of present challenges and future pathways. All collected models, benchmark datasets, source code links, and codes for evaluation have been made publicly available at https://github.com/VisionVerse/Blind-Motion-Deblurring-Survey

2401.07432

Mostafa Nouh

Mohamed Mousa and Mostafa Nouh

Parallel mechanical computing: Metamaterials that can multitask

Proceedings of the National Academy of Sciences (PNAS), 121 (52), 2024

10.1073/pnas.240743112

physics.app-ph

http://creativecommons.org/licenses/by/4.0/

Decades after being replaced with digital platforms, analogue computing has experienced a surging interest following developments in metamaterials and intricate fabrication techniques. Specifically, wave-based analogue computers which impart spatial transformations on an incident wavefront, commensurate with a desired mathematical operation, have gained traction owing to their ability to directly encode the input in its unprocessed form, bypassing analogue-to-digital conversion. While promising, these systems are inherently limited to single-task configurations. Their inability to concurrently perform multiple tasks, or compute in parallel, represents a major hindrance to advancing conceptual mechanical devices with broader computational capabilities. In here, we present a first attempt to simultaneously process independent computational tasks within the same architected structure. By breaking time invariance in a set of metasurface building blocks, multiple frequency-shifted beams are self-generated which absorb notable energy amounts from the fundamental signal. The onset of these tunable harmonics, enables distinct computational tasks to be assigned to different independent "channels", effectively allowing an analogue mechanical computer to multitask.

2401.07684

Christian Buck

N.Ackermann, H.Bonet, A.Bonhomme, C.Buck, K.F\"ulber, J.Hakenm\"uller, J.Hempfling, J.Henrichs, G.Heusser, M.Lindner, W.Maneschg, T.Rink, E.Sanchez Garcia, J.Stauber, H.Strecker and R.Wink

Final CONUS results on coherent elastic neutrino-nucleus scattering at the Brokdorf reactor

9 figures, 4 tables

CONUS Collaboration, Phys. Rev. Lett. 133, 251802 (2024)

10.1103/PhysRevLett.133.251802

hep-ex

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The Conus experiment studies coherent elastic neutrino-nucleus scattering in four 1 kg germanium spectrometers. Low ionization energy thresholds of 210 eV were achieved. The detectors were operated inside an optimized shield at the Brokdorf nuclear power plant which provided a reactor antineutrino flux of up to 2.3$\cdot$10$^{13}$\,cm$^{-2}$s$^{-1}$. In the final phase of data collection at this site, the constraints on the neutrino interaction rate were improved by an order of magnitude as compared to the previous Conus analysis. The new limit of less than 0.34 signal events kg$^{-1}$\,d$^{-1}$ is within a factor 2 of the rate predicted by the standard model. This constraint is discussed in the context of conflicting measurements and results from another reactor neutrino experiment using similar technology.

2401.08594

Kazuhiko Nishimura

Satoshi Nakano and Kazuhiko Nishimura

How do we measure trade elasticity for services?

Empirical Economics 2024

10.1007/s00181-024-02675-z

econ.GN q-fin.EC

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper is about our attempt of identifying trade elasticities through the variations in the exchange rate, for possible applications to the case of services whose physical transactions are veiled in the trade statistics. The regression analysis to estimate the elasticity entails a situation where the explanatory variable is leaked into the error term through the latent supply equation, causing an endogeneity problem for which an instrumental variable cannot be found. Our identification strategy is to utilize the normalizing condition, which enables the supply parameter to be identified, along with the reduced-form equation of the system of demand and supply equations. We evaluate the performances of the method proposed by applying to several different tangible goods, whose benchmark trade elasticities are estimable by utilizing the information on their physical transactions.

2401.09342

Omar Mustafa

Omar Mustafa, Abdullah Guvendi

Klein-Gordon oscillators in traversable wormhole rainbow gravity spacetime: Conditional exact solvability via a throat radius and oscillator frequency correlation

12 pages, 12 figures

10.1142/S0219887825500914

gr-qc

http://creativecommons.org/licenses/by/4.0/

In this study, we discuss an analytical solution for a set of the Klein-Gordon (KG) oscillators' energies through a correlation between the frequency of the KG-oscillators and the traversable wormhole (TWH) throat radius. Under such restricted parametric correlation (hence the notion of conditionally exact solvability is unavoidable in the process), we report the effects of throat radius, rainbow parameter, disclination parameter, and oscillator frequency on the spectroscopic structure of a vast number of $\left( n,m\right) $-states (the radial and magnetic quantum numbers, respectively). In the process, we only use two loop quantum gravity motivated rainbow functions pairs. The only rainbow functions that clearly and reliably fully adhere to the rainbow gravity model and secure Planck energy $E_p$ as the maximum possible energy for particles and anti-particles alike. Near the asymptotically flat upper and lower universes connected by the TWH, i.e., for the throat radius $r_{0 }>>1 $, the energies tend to cluster around the rest mass energies, i.e., $|E_{\pm }|\sim m_{0 }$. Whereas, for $r_{0 }<<1$ the energies tend to approach $|E_{\pm }| \leq E_{P}$.

2401.11141

Jian Xiao

Ji Wang, Jian Xiao, Yixuan Zou, Wenwu Xie, and Yuanwei Liu

Wideband Beamforming for RIS Assisted Near-Field Communications

in IEEE Transactions on Wireless Communications, vol. 23, no. 11, pp. 16836-16851, Nov. 2024

10.1109/TWC.2024.3447570

cs.IT eess.SP math.IT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

A near-field wideband beamforming scheme is investigated for reconfigurable intelligent surface (RIS) assisted multiple-input multiple-output (MIMO) systems, in which a deep learning-based end-to-end (E2E) optimization framework is proposed to maximize the system spectral efficiency. To deal with the near-field double beam split effect, the base station is equipped with frequency-dependent hybrid precoding architecture by introducing sub-connected true time delay (TTD) units, while two specific RIS architectures, namely true time delay-based RIS (TTD-RIS) and virtual subarray-based RIS (SA-RIS), are exploited to realize the frequency-dependent passive beamforming at the RIS. Furthermore, the efficient E2E beamforming models without explicit channel state information are proposed, which jointly exploits the uplink channel training module and the downlink wideband beamforming module. In the proposed network architecture of the E2E models, the classical communication signal processing methods, i.e., polarized filtering and sparsity transform, are leveraged to develop a signal-guided beamforming network. Numerical results show that the proposed E2E models have superior beamforming performance and robustness to conventional beamforming benchmarks. Furthermore, the tradeoff between the beamforming gain and the hardware complexity is investigated for different frequency-dependent RIS architectures, in which the TTD-RIS can achieve better spectral efficiency than the SA-RIS while requiring additional energy consumption and hardware cost.

2401.11302

Manuel Schaller

Timo Reis and Manuel Schaller

Linear-quadratic optimal control for infinite-dimensional input-state-output systems

46 pages, 6 figures

math.OC math.FA

http://creativecommons.org/licenses/by-nc-nd/4.0/

We examine the minimization of a quadratic cost functional composed of the output and the final state of abstract infinite-dimensional evolution equations in view of existence of solutions and optimality conditions. While the initial value is prescribed, we are minimizing over all inputs within a specified convex subset of square integrable controls with values in a Hilbert space. The considered class of infinite-dimensional systems is based on the system node formulation. Thus, our developed approach includes optimal control of a wide variety of linear partial differential equations with boundary control and observation that are not well-posed in the sense that the output continuously depends on the input and the initial value. We provide an application of particular optimal control problems arising in energy-optimal control of port-Hamiltonian systems. Last, we illustrate the our abstract theory by two examples including a non-well-posed heat equation with Dirichlet boundary control and a wave equation on an L-shaped domain with boundary control of the stress in normal direction.

2401.12012

Mengdi Wang

Mengdi Wang, Anna Bodonhelyi, Efe Bozkir, Enkelejda Kasneci

TurboSVM-FL: Boosting Federated Learning through SVM Aggregation for Lazy Clients

Proceedings of the AAAI Conference on Artificial Intelligence 2024 (AAAI'24)

Proceedings of the AAAI Conference on Artificial Intelligence 2024 (AAAI'24)

10.1609/aaai.v38i14.29481

cs.LG cs.DC

http://creativecommons.org/licenses/by/4.0/

Federated learning is a distributed collaborative machine learning paradigm that has gained strong momentum in recent years. In federated learning, a central server periodically coordinates models with clients and aggregates the models trained locally by clients without necessitating access to local data. Despite its potential, the implementation of federated learning continues to encounter several challenges, predominantly the slow convergence that is largely due to data heterogeneity. The slow convergence becomes particularly problematic in cross-device federated learning scenarios where clients may be strongly limited by computing power and storage space, and hence counteracting methods that induce additional computation or memory cost on the client side such as auxiliary objective terms and larger training iterations can be impractical. In this paper, we propose a novel federated aggregation strategy, TurboSVM-FL, that poses no additional computation burden on the client side and can significantly accelerate convergence for federated classification task, especially when clients are "lazy" and train their models solely for few epochs for next global aggregation. TurboSVM-FL extensively utilizes support vector machine to conduct selective aggregation and max-margin spread-out regularization on class embeddings. We evaluate TurboSVM-FL on multiple datasets including FEMNIST, CelebA, and Shakespeare using user-independent validation with non-iid data distribution. Our results show that TurboSVM-FL can significantly outperform existing popular algorithms on convergence rate and reduce communication rounds while delivering better test metrics including accuracy, F1 score, and MCC.

2401.12208

Zhihong Chen

Zhihong Chen, Maya Varma, Justin Xu, Magdalini Paschali, Dave Van Veen, Andrew Johnston, Alaa Youssef, Louis Blankemeier, Christian Bluethgen, Stephan Altmayer, Jeya Maria Jose Valanarasu, Mohamed Siddig Eltayeb Muneer, Eduardo Pontes Reis, Joseph Paul Cohen, Cameron Olsen, Tanishq Mathew Abraham, Emily B. Tsai, Christopher F. Beaulieu, Jenia Jitsev, Sergios Gatidis, Jean-Benoit Delbrouck, Akshay S. Chaudhari, Curtis P. Langlotz

A Vision-Language Foundation Model to Enhance Efficiency of Chest X-ray Interpretation

26 pages, 8 figures

cs.CV cs.CL

http://creativecommons.org/licenses/by/4.0/

Over 1.4 billion chest X-rays (CXRs) are performed annually due to their cost-effectiveness as an initial diagnostic test. This scale of radiological studies provides a significant opportunity to streamline CXR interpretation and documentation. While foundation models are a promising solution, the lack of publicly available large-scale datasets and benchmarks inhibits their iterative development and real-world evaluation. To overcome these challenges, we constructed a large-scale dataset (CheXinstruct), which we utilized to train a vision-language foundation model (CheXagent). We systematically demonstrated competitive performance across eight distinct task types on our novel evaluation benchmark (CheXbench). Beyond technical validation, we assessed the real-world utility of CheXagent in directly drafting radiology reports. Our clinical assessment with eight radiologists revealed a 36% time saving for residents using CheXagent-drafted reports, while attending radiologists showed no significant time difference editing resident-drafted or CheXagent-drafted reports. The CheXagent-drafted reports improved the writing efficiency of both radiology residents and attending radiologists in 81% and 61% of cases, respectively, without loss of quality. Overall, we demonstrate that CheXagent can effectively perform a variety of CXR interpretation tasks and holds potential to assist radiologists in routine clinical workflows.

2401.13582

Anwesh Ray

Siddhi Pathak, Anwesh Ray

Rank stability of elliptic curves in certain non-abelian extensions

Version 2: 24 pages, accepted for publication in Mathematische Nachrichten

math.NT

http://creativecommons.org/licenses/by/4.0/

Let $E_{/\mathbb{Q}}$ be an elliptic curve with rank $E(\mathbb{Q})=0$. Fix an odd prime $p$, a positive integer $n$ and a finite abelian extension $K/\mathbb{Q}$ with rank $E(K) = 0$. In this paper, we show that there exist infinitely many extensions $L/K$ such that $L/\mathbb{Q}$ is Galois with $\operatorname{Gal}(L/\mathbb{Q}) \simeq \operatorname{Gal}(K/\mathbb{Q}) \ltimes \mathbb{Z}/p^n\mathbb{Z}$, and rank $E(L)=0$. This is an extension of earlier results on rank stability of elliptic curves in cyclic extensions of prime power order to a non-abelian setting. We also obtain an asymptotic lower bound for the number of such extensions, ordered by their absolute discriminant.