text
stringlengths 133
1.92k
| summary
stringlengths 24
228
|
|---|---|
It is well-known that if R is a domain with finite character, each locally principal nonzero ideal of R is invertible. We address the problem of understanding when the converse is true and survey some recent results. | Locally principal ideals and finite character |
Supervised regression to demonstrations has been demonstrated to be a stable way to train deep policy networks. We are motivated to study how we can take full advantage of supervised loss functions for stably training deep reinforcement learning agents. This is a challenging task because it is unclear how the training data could be collected to enable policy improvement. In this work, we propose Self-Supervised Reinforcement Learning (SSRL), a simple algorithm that optimizes policies with purely supervised losses. We demonstrate that, without policy gradient or value estimation, an iterative procedure of ``labeling" data and supervised regression is sufficient to drive stable policy improvement. By selecting and imitating trajectories with high episodic rewards, SSRL is surprisingly competitive to contemporary algorithms with more stable performance and less running time, showing the potential of solving reinforcement learning with supervised learning techniques. The code is available at https://github.com/daochenzha/SSRL | Simplifying Deep Reinforcement Learning via Self-Supervision |
A new generation of laser wakefield accelerators, supported by the extreme accelerating fields generated in the interaction of PW-Class lasers and underdense targets, promises the production of high quality electron beams in short distances for multiple applications. Achieving this goal will rely heavily on numerical modeling for further understanding of the underlying physics and identification of optimal regimes, but large scale modeling of these scenarios is computationally heavy and requires efficient use of state-of-the-art Petascale supercomputing systems. We discuss the main difficulties involved in running these simulations and the new developments implemented in the OSIRIS framework to address these issues, ranging from multi-dimensional dynamic load balancing and hybrid distributed / shared memory parallelism to the vectorization of the PIC algorithm. We present the results of the OASCR Joule Metric program on the issue of large scale modeling of LWFA, demonstrating speedups of over 1 order of magnitude on the same hardware. Finally, scalability to over $\sim 10^6$ cores, and sustained performance over $\sim 2$ PFlops is demonstrated, opening the way for large scale modeling of laser wakefield accelerator scenarios. | Exploiting multi-scale parallelism for large scale numerical modelling of laser wakefield accelerators |
An N=(2|2) superfield formulation of the N=(1|1) supersymmetric Toda lattice hierarchy is proposed, and its five real forms are presented. | Hidden N=(2|2) supersymmetry of the N=(1|1) supersymmetric Toda lattice hierarchy |
All Type IIn supernovae (SNe IIn) show narrow hydrogen emission lines in their spectra. Apart from this common feature, they demonstrate very broad diversity in brightness, duration, and morphology of their light curves, which indicates that they likely come from a variety of progenitor systems and explosion channels. A particular subset of SNe IIn, the so called SNe IIn-P, exhibit $\sim$100 days plateau phases that are very similar to the ones of the ordinary hydrogen-rich SNe (SNe II). In the past, SNe IIn-P were explained by the models of sub-energetic electron capture explosions surrounded by dense extended winds. In this work, we attempt to explain this class of SNe with standard red supergiant (RSG) progenitors that experience outbursts several month before the final explosion. The outburst energies that show the best agreement between our models and the data ($5\times10^{46}\,{\rm erg}$) fall at the low range of the outburst energies that have been observed for SNe IIn (between few times $10^{46}\,{\rm erg}$ and $10^{49}\,{\rm erg}$). Instead, the inferred explosion energy of SN 2005cl is relatively high ($1-2\times10^{51}\,{\rm erg}$) compared to the explosion energies of the ordinary SNe II. Our models provide alternative explanation of SNe IIn-P to the previously proposed scenarios. | Modeling the light curve of Type IIn-P SN 2005cl with red supergiant progenitors featuring pre-SN ourbursts |
Dynamics of p53 is known to play important roles in the regulation of cell fate decisions in response to various stresses, and PDCD5 functions as a co-activator of p53 to modulate the p53 dynamics. In the present paper, we investigate how p53 dynamics are modulated by PDCD5 during the DNA damage response using methods of bifurcation analysis and potential landscape. Our results reveal that p53 activities can display rich dynamics under different PDCD5 levels, including monostability, bistability with two stable steady states, oscillations, and co-existence of a stable steady state and an oscillatory state. Physical properties of the p53 oscillations are further shown by the potential landscape, in which the potential force attracts the system state to the limit cycle attractor, and the curl flux force drives the coherent oscillation along the cyclic. We also investigate the effect of PDCD5 efficiency on inducing the p53 oscillations. We show that Hopf bifurcation is induced by increasing the PDCD5 efficiency, and the system dynamics show clear transition features in both barrier height and energy dissipation when the efficiency is close to the bifurcation point. This study provides a global picture of how PDCD5 regulates p53 dynamics via the interaction with the p53-Mdm2 oscillator and can be helpful in understanding the complicate p53 dynamics in a more complete p53 pathway. | Bifurcation analysis and potential landscape of the p53-Mdm2 oscillator regulated by the co-activator PDCD5 |
The culling process in Bootstrap Percolation is Abelian since the final stable configuration does not depend on the details of the updating procedure. An efficient algorithm is devised using this idea for the determination of the bootstrap percolation threshold in two dimension which takes $L^2$ time compared to the $L^3 \log L$ in the conventional method. A generalised Bootstrap Percolation allowing many particles at a site is studied where continuous phase transitions are observed for all values of the threshold parameter. Similar results are also obtained for the continuum Bootstrap Percolation model. | Abelian Cascade Dynamics in Bootstrap Percolation |
One of the most well-established applications of machine learning is in deciding what content to show website visitors. When observation data comes from high-velocity, user-generated data streams, machine learning methods perform a balancing act between model complexity, training time, and computational costs. Furthermore, when model freshness is critical, the training of models becomes time-constrained. Parallelized batch offline training, although horizontally scalable, is often not time-considerate or cost-effective. In this paper, we propose Lambda Learner, a new framework for training models by incremental updates in response to mini-batches from data streams. We show that the resulting model of our framework closely estimates a periodically updated model trained on offline data and outperforms it when model updates are time-sensitive. We provide theoretical proof that the incremental learning updates improve the loss-function over a stale batch model. We present a large-scale deployment on the sponsored content platform for a large social network, serving hundreds of millions of users across different channels (e.g., desktop, mobile). We address challenges and complexities from both algorithms and infrastructure perspectives, and illustrate the system details for computation, storage, and streaming production of training data. | Lambda Learner: Fast Incremental Learning on Data Streams |
For a reductive group $G$ we equip the category of $G_\mathcal{O}$-equivariant polarizable pure Hodge modules on the affine Grassmannian $\mathrm{Gr}_G$ with a structure of neutral Tannakian category. We show that it is equivalent to a twisted tensor product of the category of representations of the Langlands dual group and the category of pure polarizable Hodge structures. | Satake equivalence for Hodge modules on affine Grassmannians |
Si-SiOx superlattices are among the candidates that have been proposed as high band gap absorber material in all-Si tandem solar cell devices. Due to the large potential barriers for photoexited charge carriers, transport in these devices is restricted to quantum confined superlattice states. As a consequence of the finite number of wells, large built-in fields and any kind of disorder, the electronic spectrum can deviate considerably from the minibands of a regular superlattice. In this paper, a quantum-kinetic theory based on the non-equilibrium Green's function formalism for an effective mass Hamiltonian is used to investigate photogeneration and transport in such devices for arbitrary geometry and operating conditions. By including the coupling of electrons to both photons and phonons, the theory is able to provide a microscopic picture of indirect generation, carrier relaxation and inter-well transport mechanisms beyond the ballistic regime. | Theory and simulation of photogeneration and transport in Si-SiOx superlattice absorbers |
Solotronics, optoelectronics based on solitary dopants, is an emerging field of research and technology reaching the ultimate limit of miniaturization. It aims at exploiting quantum properties of individual ions or defects embedded in a semiconductor matrix. As already shown, optical control of a spin of a magnetic ion is feasible employing photo-generated carriers confined in a quantum dot. A non-radiative recombination, regarded as a severe problem, limited development of quantum dots with magnetic ions. Our photoluminescence studies on, so far unexplored, individual CdTe dots with single cobalt ions and individual CdSe dots with single manganese ions show, however, that even if energetically allowed, the single ion related non-radiative recombination is negligible in such zero-dimensional structures. This opens solotronics for a wide range of even not yet considered systems. Basing on the results of our single spin relaxation experiments and on the material trends, we identify optimal magnetic ion-quantum dot systems for implementation of a single-ion based spin memory. | Designing quantum dots for solotronics |
In this paper we consider the use of Gamma Ray Bursts (GRBs) as distance markers to study the unification of dark energy and dark matter in the context of the so-called Generalized Chaplygin Gas (GCG) model. We consider that the GRB luminosity may be estimated from its variability, time-lag, and also use the so-called Ghirlanda relation. We evaluate the improvements one may expect once more GRBs and their redshift become available. We show that although GRBs allow for extending the Hubble diagram to higher redshifts, its use as a dark energy probe is limited when compared to SNe Ia. We find that the information from GRBs can provide some bounds on the amount of dark matter and dark energy independently of the equation of state. This is particularly evident for XCDM-type models, which are, for low-redshifts ($z\leq2$), degenerate with the GCG. | Gamma-ray bursts as dark energy-matter probes in the context of the generalized Chaplygin gas model |
We report benchmark calculations of electroabsorption in semiconducting single-walled carbon nanotubes to provide motivation to experimentalists to perform electroabsorption measurement on these systems. We show that electroabsorption can detect continuum bands in different energy manifolds, even as other nonlinear absorption measurements have failed to detect them. Direct determination of the binding energies of excitons in higher manifolds thereby becomes possible. We also find that electroabsorption can provide evidence for Fano-type coupling between the second exciton and the lowest continuum band states. | Elucidation of the electronic structure of semiconducting single-walled carbon nanotubes by electroabsorption spectroscopy |
The invariant differential cross section of inclusive $\omega(782)$ meson production at midrapidity ($|y|<0.5$) in pp collisions at $\sqrt{s}$ = 7 TeV was measured with the ALICE detector at the LHC over a transverse momentum range of 2 < $p_{\rm{T}}$ < 17 GeV/$c$. The $\omega$ meson was reconstructed via its $\omega\rightarrow\pi^+\pi^-\pi^0$ decay channel. The measured $\omega$ production cross section is compared to various calculations: PYTHIA 8.2 Monash 2013 describes the data, while PYTHIA 8.2 Tune 4C overestimates the data by about 50%. A recent NLO calculation, which includes a model describing the fragmentation of the whole vector-meson nonet, describes the data within uncertainties below 6 GeV/$c$, while it overestimates the data by up to 50% for higher $p_{\rm{T}}$. The $\omega/\pi^0$ ratio is in agreement with previous measurements at lower collision energies and the PYTHIA calculations. In addition, the measurement is compatible with transverse mass scaling within the measured $p_{\rm{T}}$ range and the ratio is constant with $C^{\omega/\pi^{0}}$ = 0.67 $\pm$ 0.03 (stat) $\pm$ 0.04 (sys) above a transverse momentum of 2.5 GeV/$c$. | Production of $\omega$ mesons in pp collisions at $\sqrt{s}$ = 7 TeV |
Learning disentangled representations from visual data, where different high-level generative factors are independently encoded, is of importance for many computer vision tasks. Solving this problem, however, typically requires to explicitly label all the factors of interest in training images. To alleviate the annotation cost, we introduce a learning setting which we refer to as "reference-based disentangling". Given a pool of unlabeled images, the goal is to learn a representation where a set of target factors are disentangled from others. The only supervision comes from an auxiliary "reference set" containing images where the factors of interest are constant. In order to address this problem, we propose reference-based variational autoencoders, a novel deep generative model designed to exploit the weak-supervision provided by the reference set. By addressing tasks such as feature learning, conditional image generation or attribute transfer, we validate the ability of the proposed model to learn disentangled representations from this minimal form of supervision. | Learning Disentangled Representations with Reference-Based Variational Autoencoders |
Let $G$ be graph on $n$ vertices and $G^{(t)}$ its blow-up graph of order $t.$ In this paper, we determine all eigenvalues of the Laplacian and the signless Laplacian matrix of $G^{(t)}$ and its complement $\bar{G^{(t)}}.$ | Spectra of the blow-up graphs |
Based on an optical analogy of spintronics, the generation of optical tornadoes is theoretically investigated in two-dimensional photonic crystals without space-inversion symmetry. We address its close relation to the Berry curvature in crystal momentum space, which represents the non-trivial geometric property of a Bloch state. It is shown that the Berry curvature is easily controlled by tuning two types of dielectric rods in a honeycomb photonic crystal. Then, Bloch states with large Berry curvatures appear as optical tornadoes in real space. The radiation force of such a configuration is analyzed, and its possible application is also discussed. | Optical tornadoes in photonic crystals |
(This is a general physics level overview article about hidden sectors, and how they motivate searches for long-lived particles. Intended for publication in Physics Today.) Searches for new physics at the Large Hadron Collider have so far come up empty, but we just might not be looking in the right place. Spectacular bursts of particles appearing seemingly out of nowhere could shed light on some of nature's most profound mysteries. | Flashes of Hidden Worlds at Colliders |
We perform a proof-of-concept implementation of the massively parallel algorithm (P.M. Lushnikov, Opt. Lett., v. 27, 939 (2002)) for simulation of dispersion-managed wavelength-division-multiplexed optical fiber systems. Linear scalability of the algorithm with the number of computer cores is demonstrated. Exact result on the accuracy of the implemented algorithm is found analytically and confirmed numerically as well as it is compared with to accuracy of the standard split-step algorithm. | Proof of concept implementation of the massively parallel algorithm for simulation of dispersion-managed WDM optical fiber systems |
We investigate the phase structure of two-color QCD at both real and imaginary chemical potentials mu, performing lattice simulations and analyzing the data with the Polyakov-loop extended Nambu--Jona-Lasinio (PNJL) model. Lattice QCD simulations are done on an 8^3 times 4 lattice with the clover-improved two-flavor Wilson fermion action and the renormalization-group improved Iwasaki gauge action. We test the analytic continuation of physical quantities from imaginary mu to real mu by comparing lattice QCD results calculated at real mu with the result of analytic function the coefficients of which are determined from lattice QCD results at imaginary mu. We also test the validity of the PNJL model by comparing model results with lattice QCD ones. The PNJL model is good in the deconfinement region, but less accurate in the transition and confinement regions. This problem is improved by introducing the baryon degree of freedom to the model. It is also found that the vector-type four-quark interaction is necessary to explain lattice data on the quark number density. | Phase structure of two-color QCD at real and imaginary chemical potentials; lattice simulations and model analyses |
Hybrid Josephson junctions with three or more superconducting terminals coupled to a semiconducting region are expected to exhibit a rich variety of phenomena. Andreev bound states (ABSs) arising in such multiterminal devices have been predicted to form unconventional energy band structures, which can be engineered by controlling superconducting phase differences between the individual terminals. Here we report tunnelling spectroscopy measurements of three-terminal Josephson junctions (3TJJs) realised in a gate-tunable InAs/Al heterostructure. The three terminals are connected to form two loops where phase differences are independently controlled. Our results provide a demonstration of phase control over a novel Andreev band structure. Signatures of hybridisation between ABSs, consistent with their overlap in the 3TJJ region, are observed in the form of avoided crossings in the spectrum and are well explained by a numerical model. Future extensions of this work could focus on addressing spin-resolved energy levels, ground state parity transitions and Weyl bands in multiterminal geometries. | Hybridisation of Andreev bound states in three-terminal Josephson junctions |
We present an agent-based model inspired by the Evolutionary Minority Game (EMG), albeit strongly adapted to the case of competition for limited resources in ecology. The agents in this game become able, after some time, to predict the a priori best option as a result of an evolution-driven learning process. We show that a self-segregated social structure can emerge from this process, i.e., extreme learning strategies are always favoured while intermediate learning strategies tend to die out. This result may contribute to understanding some levels of organization and cooperative behaviour in ecological and social systems. We use the ideas and results reported here to discuss an issue of current interest in ecology: the mistimings in egg laying observed for some species of bird as a consequence of their slower rate of adaptation to climate change in comparison with that shown by their prey. Our model supports the hypothesis that habitat-specific constraints could explain why different populations are adapting differently to this situation, in agreement with recent experiments. | Limited resources and evolutionary learning may help to understand the mistimed reproduction in birds caused by climate change |
The hourglass fermions in solid-state materials have been attracting significant interest recently. However, realistic two-dimensional (2D) materials with hourglass-shaped band structures are still very scarce. Here, through the first-principles calculations, we identify the monolayer Pb(ClO$_{2}$)$_{2}$ and Sr(ClO$_{2}$)$_{2}$ materials as the new realistic materials platform to realize 2D hourglass Weyl nodal loop. We show that these monolayer materials possess an hourglass Weyl nodal loop circling around the $\Gamma$ point and Weyl nodal line on the Brillouin zone (BZ) boundary in the absence of spin-orbit coupling (SOC). Through the symmetry analysis, we demonstrate that the hourglass Weyl nodal loop and Weyl nodal line are protected by the nonsymmorphic symmetries, and are robust under the biaxial strains. When we include the SOC, a tiny gap will be opened in the hourglass nodal loop and nodal line, and the nodal line can be transformed into the spin-orbit Dirac points. Our results provide a new realistic material platform for studying the intriguing physics associated with the 2D hourglass Weyl nodal loop and spin-orbit Dirac points. | Two-dimensional hourglass Weyl nodal loop in monolayer Pb(ClO$_{2}$)$_{2}$ and Sr(ClO$_{2}$)$_{2}$ |
We provide a representation for the scaling limit of the d=2 critical Ising magnetization field as a (conformal) random field using SLE (Schramm-Loewner Evolution) clusters and associated renormalized area measures. The renormalized areas are from the scaling limit of the critical FK (Fortuin-Kasteleyn) clusters and the random field is a convergent sum of the area measures with random signs. Extensions to off-critical scaling limits, to d=3 and to Potts models are also considered. | Ising (Conformal) Fields and Cluster Area Measures |
We prove the existence and uniqueness of Stochastic Lagrangian Flows and almost everywhere Stochastic Flows for non-degenearted SDEs with rough coefficients. As an application of our main result, we show that there exists a unique Stochastic Flow corresponding to each Leray-Hopf solution of 3D Navier-Stokes equation in the DiPerna-Lions sense. | Stochastic Lagrangian Flows for SDEs with rough coefficients |
We provide strong evidence that the relaxation dynamics of one-dimensional, metallic Fermi systems resulting out of an abrupt amplitude change of the two-particle interaction has aspects which are universal in the Luttinger liquid sense: The leading long-time behavior of certain observables is described by universal functions of the equilibrium Luttinger liquid parameter and the renormalized velocity. We analytically derive those functions for the Tomonaga-Luttinger model and verify our hypothesis of universality by considering spinless lattice fermions within the framework of the density matrix renormalization group. | Luttinger liquid universality in the time evolution after an interaction quench |
The number of works addressing the role of energy efficiency in the software development has been increasing recently. But, designers and programmers still complain about the lack of tools that help them to make energy-efficiency decisions. Some works show that energy-aware design decisions tend to have a larger impact in the power consumed by applications, than code optimizations. In this paper we present the HADAS green assistant, which helps developers to identify the energy-consuming concerns of their applications (i.e., points in the application that consume more energy, like storing or transferring data), and also to model, analyse and reason about different architectural solutions for each of these concerns. This tool models the variability of more or less green architectural practices and the dependencies between different energy-consuming concerns using variabilty models. Finally, this tool will automatically generate the architectural configuration derived from the selections made by the developer from an energy consumption point of view. | HADAS Green Assistant: designing energy-efficient applications |
Fast spin manipulation in magnetic heterostructures, where magnetic interactions between different materials often define the functionality of devices, is a key issue in the development of ultrafast spintronics. Although recently developed optical approaches such as ultrafast spin-transfer and spin-orbit torques open new pathways to fast spin manipulation, these processes do not fully utilize the unique possibilities offered by interfacial magnetic coupling effects in ferromagnetic multilayer systems. Here, we experimentally demonstrate ultrafast photo-enhanced interfacial exchange interactions in the ferromagnetic Co$_2$FeAl/(Ga,Mn)As system at low laser fluence levels. The excitation efficiency of Co$_2$FeAl with the (Ga,Mn)As layer is 30-40 times higher than the case with the GaAs layer at 5 K due to a photo-enhanced exchange coupling interaction via photoexcited charge transfer between the two ferromagnetic layers. In addition, the coherent spin precessions persist to room temperature, excluding the drive of photo-enhanced magnetization in the (Ga,Mn)As layer and indicating a proximity-effect-related optical excitation mechanism. The results highlight the importance of considering the range of interfacial exchange interactions in ferromagnetic heterostructures and how these magnetic coupling effects can be utilized for ultrafast, low-power spin manipulation. | Ultrafast enhancement of interfacial exchange coupling in ferromagnetic bilayer |
We present an analytical model for jets in Fanaroff & Riley Class I (FRI) radio galaxies, in which an initially laminar, relativistic flow is surrounded by a shear layer. We apply the appropriate conservation laws to constrain the jet parameters, starting the model where the radio emission is observed to brighten abruptly. We assume that the laminar flow fills the jet there and that pressure balance with the surroundings is maintained from that point outwards. Entrainment continuously injects new material into the jet and forms a shear layer, which contains material from both the environment and the laminar core. The shear layer expands rapidly with distance until finally the core disappears, and all of the material is mixed into the shear layer. Beyond this point, the shear layer expands in a cone and decelerates smoothly. We apply our model to the well-observed FRI source 3C31 and show that there is a self-consistent solution. We derive the jet power, together with the variations of mass flux and and entrainment rate with distance from the nucleus. The predicted variation of bulk velocity with distance in the outer parts of the jets is in good agreement with model fits to VLA observations. Our prediction for the shape of the laminar core can be tested with higher-resolution imaging. | A relativistic mixing-layer model for jets in low-luminosity radio galaxies |
Isobar collisions, $^{96}_{44}$Ru+$^{96}_{44}$Ru and $^{96}_{40}$Zr+$^{96}_{40}$Zr, at $\sqrt{s_{\mathrm {NN}}}$ = 200 GeV have been performed at RHIC in order to study the charge separation along the magnetic field, called the Chiral Magnetic Effect (CME). The difference in nuclear deformation and structure between the two isobar nuclei may result in a difference in the flow magnitudes. Hence, elliptic flow measurements for these collisions give direct information about the initial state anisotropies. Strange and multi-strange hadrons have a small hadronic cross-section compared to light hadrons, making them an excellent probe for understanding the initial state anisotropies of the medium produced in these isobar collisions. The collected datasets include approximately two billion events for each of the isobar species and provide a unique opportunity for statistics hungry measurements. In this proceeding, we will report the elliptic flow ($v_{2}$) measurement of $K_{s}^{0}$, $\Lambda$, $\overline{\Lambda}$, $\phi$, $\Xi^{-}$, $\overline{\Xi}^{+}$, $\Omega^{-}$, and $\overline{\Omega}^{+}$ at mid-rapidity for Ru+Ru and Zr+Zr collisions at $\sqrt{s_{\mathrm {NN}}}$ = 200 GeV. The centrality and transverse momentum ($p_{T}$) dependence of $v_{2}$ of (multi-)strange hadrons will be shown. System size dependence of $v_{2}$ will be shown by comparing the $v_{2}$ results obtained from Cu+Cu, Au+Au, and U+U collisions. The number of constituent quark (NCQ) scaling for these strange hadrons will also be tested. We will also compare the $p_{T}$-integrated $v_{2}$ for these two isobar collisions. Transport model calculations will be compared to data to provide further quantitative constraints on the nuclear structure. | Probing nuclear structure using elliptic flow of strange and multi-strange hadrons in isobar collisions |
Ordering of the geometrically frustrated two-dimensional Heisenberg antiferromagnet on a pyrochlore slab is studied by Monte Carlo simulations. In contrast to the kagom\'e Heisenberg antiferromagnet, the model exhibits locally non-coplanar spin structures at low temperatures, bearing nontrivial chiral degrees of freedom. Under certain conditions, the model exhibits a novel Kosterlitz-Thouless-type transition at a finite temperature associated with these chiral degrees of freedom. | Chiral Kosterlitz-Thouless transition in the frustrated Heisenberg antiferromagnet on a pyrochlore slab |
Intent detection is a crucial task in any Natural Language Understanding (NLU) system and forms the foundation of a task-oriented dialogue system. To build high-quality real-world conversational solutions for edge devices, there is a need for deploying intent detection model on device. This necessitates a light-weight, fast, and accurate model that can perform efficiently in a resource-constrained environment. To this end, we propose LIDSNet, a novel lightweight on-device intent detection model, which accurately predicts the message intent by utilizing a Deep Siamese Network for learning better sentence representations. We use character-level features to enrich the sentence-level representations and empirically demonstrate the advantage of transfer learning by utilizing pre-trained embeddings. Furthermore, to investigate the efficacy of the modules in our architecture, we conduct an ablation study and arrive at our optimal model. Experimental results prove that LIDSNet achieves state-of-the-art competitive accuracy of 98.00% and 95.97% on SNIPS and ATIS public datasets respectively, with under 0.59M parameters. We further benchmark LIDSNet against fine-tuned BERTs and show that our model is at least 41x lighter and 30x faster during inference than MobileBERT on Samsung Galaxy S20 device, justifying its efficiency on resource-constrained edge devices. | LIDSNet: A Lightweight on-device Intent Detection model using Deep Siamese Network |
In this paper we show that the empirical eigenvalue distribution of any sample covariance matrix generated by independent copies of a stationary regular sequence has a limiting distribution depending only on the spectral density of the sequence. We characterize this limit in terms of Stieltjes transform via a certain simple equation. No rate of convergence to zero of the covariances is imposed. If the entries of the stationary sequence are functions of independent random variables the result holds without any other additional assumptions. As a method of proof, we study the empirical eigenvalue distribution for a symmetric matrix with independent rows below the diagonal; the entries satisfy a Lindeberg-type condition along with mixingale-type conditions without rates. In this nonstationary setting we point out a property of universality, meaning that, for large matrix size, the empirical eigenvalue distribution depends only on the covariance structure of the sequence and is independent on the distribution leading to it. These results have interest in themselves, allowing to study symmetric random matrices generated by random processes with both short and long memory. | On the empirical spectral distribution for matrices with long memory and independent rows |
Acoustic and linguistic analysis for elderly emotion recognition is an under-studied and challenging research direction, but essential for the creation of digital assistants for the elderly, as well as unobtrusive telemonitoring of elderly in their residences for mental healthcare purposes. This paper presents our contribution to the INTERSPEECH 2020 Computational Paralinguistics Challenge (ComParE) - Elderly Emotion Sub-Challenge, which is comprised of two ternary classification tasks for arousal and valence recognition. We propose a bi-modal framework, where these tasks are modeled using state-of-the-art acoustic and linguistic features, respectively. In this study, we demonstrate that exploiting task-specific dictionaries and resources can boost the performance of linguistic models, when the amount of labeled data is small. Observing a high mismatch between development and test set performances of various models, we also propose alternative training and decision fusion strategies to better estimate and improve the generalization performance. | Is Everything Fine, Grandma? Acoustic and Linguistic Modeling for Robust Elderly Speech Emotion Recognition |
We calculate the part of the order $\alpha_s^2$ correction to the semileptonic heavy quark decay rate proportional to the number of light quark flavors, and use our result to set the scale for evaluating the strong coupling in the order $\alpha_s$ term according to the scheme of Brodsky, Lepage and Mackenzie. Expressing the decay rate in terms of the heavy quark pole mass $m_Q$, we find the scale for the $\overline{MS}$ strong coupling to be $0.07\, m_Q$. If the decay rate is expressed in terms of the $\overline{MS}$ heavy quark mass $\overline m_Q(m_Q)$ then the scale is $0.12\, m_Q$. We use these results along with the existing calculations for hadronic $\tau$ decay to calculate the BLM scale for the nonleptonic decay width and the semileptonic branching ratio. The implications for the value of $|V_{bc}|$ extracted from the inclusive semileptonic $B$ meson decay rate are discussed. | Perturbative Strong Interaction Corrections to the Heavy Quark Semileptonic Decay Rate |
Landau level spectroscopy has been employed to probe the electronic structure of the valence band in a series of p-type HgTe/HgCdTe quantum wells with both normal and inverted ordering of bands. We find that the standard axial-symmetric 4-band Kane model, which is nowadays widely applied in physics of HgTe-based topological materials, does not fully account for the complex magneto-optical response observed in our experiments - notably, for the unexpected avoided crossings of excitations and for the appearance of transitions that are electric-dipole forbidden within this model. Nevertheless, reasonable agreement with experiments is achieved when the standard model is expanded to include effects of bulk and interface inversion asymmetries. These remove the axial symmetry, and among other, profoundly modify the shape of valence bands. | Landau level spectroscopy of valence bands in HgTe quantum wells: Effects of symmetry lowering |
The purpose of this paper is to derive sharp asymptotics of the ground state energy for the heavy atoms and molecules in the relativistic settings, with the self-generated magnetic field, and, in particular, to derive relativistic Scott correction term and also Dirac, Schwinger and relativistic correction terms. Also we will prove that Thomas-Fermi density approximates the actual density of the ground state, which opens the way to estimate the excessive negative and positive charges and the ionization energy. | Asymptotics of the ground state energy in the relativistic settings and with self-generated magnetic field |
High-order harmonic generation (HHG) from molecular ion HeH$^{2+}$ in two initial states of $1s\sigma$ (ground state) and $2p\sigma$ (first excited state) is investigated theoretically, in homogeneous and plasmonic-enhanced laser fields. The electron ionization and the enhancement of HHG yield in ground and first excited states, as the initial states, is studied and a strong orientation effect for ionization of electron in $2p\sigma$ state is discovered, that it can be used to reach a longer lifetime for the electron in the first excited state, which is responsible for the enhancement in HHG yield, during the high harmonic emission. In order to investigate effect of plasmonic field on the cutoff position of HHG spectra, first the enhancement of laser field in a nano-antennae is calculated and then the interaction of the enhanced field with HeH$^{2+}$ molecule is studied by solving the time-dependent Schr\"{o}dinger equation. According to the results, when HeH$^{2+}$ molecule, in the initial state of $2p\sigma$, is irradiated by plasmonic polarized laser field, with the polarization normal to the molecular axis, the HHG yield and cutoff position enhance while the excited state has a long lifetime during the HHG process. | Quantum mechanical study of high-order harmonic generation from HeH$^{2+}$ molecule in homogeneous and plasmonic-enhanced laser Fields |
In this paper, we present the testing of four hypotheses on two streams of observations that are driven by L\'evy processes. This is applicable for sequential decision making on the state of two-sensor systems. In one case, each sensor receives or does not receive a signal obstructed by noise. In another, each sensor receives data-driven by L\'evy processes with large or small jumps. In either case, these give rise to four possibilities. Infinitesimal generators are presented and analyzed. Bounds for infinitesimal generators in terms of \emph{super-solutions} and \emph{sub-solutions} are computed. An application of this procedure for the stochastic model is also presented in relation to the financial market. | Infinitesimal generators for two-dimensional L\'evy process-driven hypothesis testing |
A cohomology theory for "odd polygon" relations -- algebraic imitations of Pachner moves in dimensions 3, 5, ... -- is constructed. Manifold invariants based on polygon relations and nontrivial polygon cocycles are proposed. Example calculation results are presented. | Odd-gon relations and their cohomology |
White matter fiber clustering is an important strategy for white matter parcellation, which enables quantitative analysis of brain connections in health and disease. In combination with expert neuroanatomical labeling, data-driven white matter fiber clustering is a powerful tool for creating atlases that can model white matter anatomy across individuals. While widely used fiber clustering approaches have shown good performance using classical unsupervised machine learning techniques, recent advances in deep learning reveal a promising direction toward fast and effective fiber clustering. In this work, we propose a novel deep learning framework for white matter fiber clustering, Deep Fiber Clustering (DFC), which solves the unsupervised clustering problem as a self-supervised learning task with a domain-specific pretext task to predict pairwise fiber distances. This process learns a high-dimensional embedding feature representation for each fiber, regardless of the order of fiber points reconstructed during tractography. We design a novel network architecture that represents input fibers as point clouds and allows the incorporation of additional sources of input information from gray matter parcellation to improve anatomical coherence of clusters. In addition, DFC conducts outlier removal naturally by rejecting fibers with low cluster assignment probability. We evaluate DFC on three independently acquired cohorts, including data from 220 individuals across genders, ages (young and elderly adults), and different health conditions (healthy control and multiple neuropsychiatric disorders). We compare DFC to several state-of-the-art white matter fiber clustering algorithms. Experimental results demonstrate superior performance of DFC in terms of cluster compactness, generalization ability, anatomical coherence, and computational efficiency. | Deep fiber clustering: Anatomically informed fiber clustering with self-supervised deep learning for fast and effective tractography parcellation |
We report the results of our Chandra observations of six QSOs at $z\sim 3$ from the Palomer Transit Grism Survey. Our primary goal is to investigate the possible systematic change of $\alpha_{ox}$ between $z>4$ and $z\sim 3$, between which a rapid rise of luminous QSO number density with cosmic time is observed. The summed spectrum showed a power-law spectrum with photon index of $\Gamma \approx 1.9$, which is similar to other unabsorbed AGNs. Combining our $z\sim 3$ QSOs with X-ray observations of QSOs at $z>4$ from literaure/archive, we find a correlation of $\alpha_{\rm ox}$ with optical luminosity. This is consistent with the fact that the luminosity function slope of the luminous end of the X-ray selected QSOs is steeper than that of optically-selected QSOs. We discuss an upper limit to the redshift dependence of $\alpha_{ox}$ using a Monte-Carlo simulation. Within the current statistical errors including the derived limits on the redshift dependence of $\alpha_{\rm ox}$, we found that the behaviors of the X-ray and optically-selected QSO number densities are consistent with each other. | Chandra Observations of Six QSOs at z $\approx$ 3 |
Recently, interest has grown for application of reconfigurable devices in robust and adaptive control systems. The main advantage of such devices is that its structure is not fixed and may be varied depending on the currently used control algorithm. In this paper a new PC-based reconfigurable microcontrollers (RMC) are offered for Experimental Physics control systems. | PLD-Based Reconfigurable Controllers for Feedback Systems |
We probe the evolution of the fine-structure constant, alpha, with cosmic time. Accurate positions of the FeII lines 1608, 2382,and 2600 A are measured in the z = 1.84 absorption system from a high-resolution (FWHM = 3.8 km/s) and high signal-to-noise (S/N >= 100) spectrum of the quasar Q1101-264 (z_em = 2.15, V = 16.0), integrated for 15.4 hours. The Single Ion Differential alpha Measurement (SIDAM) procedure and the Delta chi^2 method are used to set constraints on Delta alpha/alpha. We have found a relative radial velocity shift between the 1608 A and 2382,2600 A lines of Delta v = -180 +/- 85 m/s (both random and systematic errors are included), which, if real, would correspond to Delta alpha/alpha = (5.4 +/- 2.5) 10^{-6} (1sigma C.L.). Considering the strong implications of a such variability, additional observations with comparable accuracy at redshift z ~ 1.8 are required to confirm this result. | A new measure of Delta alpha/alpha at redshift z = 1.84 from very high resolution spectra of Q1101-264 |
We propose a time-independent method for finding a correlated ground state of an extended time-dependent Hartree-Fock theory, known as the time-dependent density-matrix theory (TDDM). The correlated ground state is used to formulate the small amplitude limit of TDDM (STDDM) which is a version of extended RPA theories with ground-state correlations. To demonstrate the feasibility of the method, we calculate the ground state of 22O and study the first 2+ state and its two-phonon states using STDDM. | Extended RPA with ground-state correlations |
We prove that $X^r$ follows an FID distribution if: (1) $X$ follows a free Poisson distribution without an atom at 0 and $r\in(-\infty,0]\cup[1,\infty)$; (2) $X$ follows a free Poisson distribution with an atom at 0 and $r\geq1$; (3) $X$ follows a mixture of some HCM distributions and $|r|\geq1$; (4) $X$ follows some beta distributions and $r$ is taken from some interval. In particular, if $S$ is a standard semicircular element then $|S|^r$ is freely infinitely divisible for $r\in(-\infty,0]\cup[2,\infty)$. Also we consider the symmetrization of the above probability measures, and in particular show that $|S|^r \,\text{sign}(S)$ is freely infinitely divisible for $r\geq2$. Therefore $S^n$ is freely infinitely divisible for every $n\in\mathbb N$. The results on free Poisson and semicircular random variables have a good correspondence with classical ID properties of powers of gamma and normal random variables. | Free infinite divisibility for powers of random variables |
The Coifman-Fefferman inequality implies quite easily that a Calderon-Zygmund operator $T$ acts boundedly in a Banach lattice $X$ on $\mathbb R^n$ if the Hardy-Littlewood maximal operator $M$ is bounded in both $X$ and $X'$. We discuss this phenomenon in some detail and establish a converse result under the assumption that $X$ is $p$-convex and $q$-concave with some $1 < p, q < \infty$ and satisfies the Fatou property: if a linear operator $T$ is bounded in $X$ and $T$ is nondegenerate in a certain sense (for example, if $T$ is a Riesz transform) then $M$ has to be bounded in both $X$ and $X'$. | A_1-regularity and boundedness of Calderon-Zygmund operators |
We present a study of the behavior of two different figures of merit for quantum correlations, entanglement of formation and quantum discord, under quantum channels showing how the former can, counterintuitively, be more resilient to such environments spoiling effects. By exploiting strict conservation relations between the two measures and imposing necessary constraints on the initial conditions we are able to explicitly show this predominance is related to build-up of the system-environment correlations. | Predominance of entanglement of formation over quantum discord under quantum channels |
The structural characteristics of the perovskite- based ferroelectric Pb(Zn1/3Nb2/3)O3-9%PbTiO3 at the morphotropic phase boundary (MPB) region (x≃0.09) have been analyzed. The analysis is based on the symmetry adapted free energy functions under the assumption that the total polarization and the unit cell volume are conserved during the transformations between various morphotropic phases. Overall features of the relationships between the observed lattice constants at various conditions have been consistently explained. The origin of the anomalous physical properties at MPB is discussed. | Symmetry of high-piezoelectric Pb-based complex perovskites at the morphotropic phase boundary II. Theoretical treatment |
Angle-resolved soft x-ray measurements made at the boron K-edge in single crystal MgB2 provide new insights into the B-2p local partial density of both unoccupied and occupied band states. The strong variation of absorption with incident angle of exciting x-rays permits the clear separation of contributions from \sigma states in the boron plane and \pi states normal to the plane. A careful comparison with theory accurately determines the energy of selected critical $k$ points in the conduction band. Resonant inelastic x-ray emission at an incident angle of 15 degrees shows a large enhancement of the emission spectra within about 0.5 eV of the Fermi level that is absent at 45 degrees and is much reduced at 60 degrees. We conclude that momentum transferred from the resonant inelastic x-ray scattering (RIXS) process couples empty and filled states across the Fermi level. | An angle-resolved soft x-ray spectroscopy study of the electronic states of single crystal MgB2 |
A theorem of Lawson and Simons states that the only stable minimal submanifolds in complex projective spaces are complex submanifolds. We generalize their result to the cases of quaternionic and octonionic projective spaces. Our approach gives a unified viewpoint towards conformal and projective geometries. | Conformal geometry and special holonomy |
The method of analytic continuation from imaginary to real chemical potential is one of the most powerful tools to circumvent the sign problem in lattice QCD. Here we test this method in a theory, 2-color QCD, which is free from the sign problem. We find that the method gives reliable results, within appropriate ranges of the chemical potential, and that a considerable improvement can be achieved if suitable functions are used to interpolate data with imaginary chemical potential. | Analytic continuation from imaginary to real chemical potential in two-color QCD |
Video summarization aims to select representative frames to retain high-level information, which is usually solved by predicting the segment-wise importance score via a softmax function. However, softmax function suffers in retaining high-rank representations for complex visual or sequential information, which is known as the Softmax Bottleneck problem. In this paper, we propose a novel framework named Dual Mixture Attention (DMASum) model with Meta Learning for video summarization that tackles the softmax bottleneck problem, where the Mixture of Attention layer (MoA) effectively increases the model capacity by employing twice self-query attention that can capture the second-order changes in addition to the initial query-key attention, and a novel Single Frame Meta Learning rule is then introduced to achieve more generalization to small datasets with limited training sources. Furthermore, the DMASum significantly exploits both visual and sequential attention that connects local key-frame and global attention in an accumulative way. We adopt the new evaluation protocol on two public datasets, SumMe, and TVSum. Both qualitative and quantitative experiments manifest significant improvements over the state-of-the-art methods. | Query Twice: Dual Mixture Attention Meta Learning for Video Summarization |
Within the recently proposed doped-carrier representation of the projected lattice electron operators we derive a full Ising version of the t-J model. This model possesses the global discrete Z_2 symmetry as a maximal spin symmetry of the Hamiltonian at any values of the coupling constants, t and J. In contrast, in the spin anisotropic limit of the t-J model, usually referred to as the t-J_z model, the global SU(2) invariance is fully restored at J_z=0, so that only the spin-spin interaction has in that model the true Ising form. We discuss a relationship between those two models and the standard isotropic t-J model. We show that the low-energy quasiparticles in all three models share the qualitatively similar properties at low doping and small values of J/t. The main advantage of the proposed Ising t-J model over the t-J_z one is that the former allows for the unbiased Monte Carlo calculations on large clusters of up to 10^3 sites. Within this model we discuss in detail the destruction of the antiferromagnetic order by doping as well as the interplay between the AF order and hole mobility. We also discuss the effect of the exchange interaction and that of the next nearest neighbour hoppings on the destruction of the AF order at finite doping. We show that the short-range AF order is observed in a wide range of temperatures and dopings, much beyond the boundaries of the AF phase. We explicitly demonstrate that the local no double occupancy constraint plays the dominant role in destroying the magnetic order at finite doping. Finally, a role of inhomogeneities is discussed. | Ising t-J model close to half filling: A Monte Carlo study |
We report the electronic characterization of mesoscopic Hall bar devices fabricated from coupled InAs/GaSb quantum wells sandwiched between AlSb barriers, an emerging candidate for two-dimensional topological insulators. The electronic width of the etched structures was determined from the low field magneto-resistance peak, a characteristic signature of partially diffusive boundary scattering in the ballistic limit. In case of dry-etching the electronic width was found to decrease with electron density. In contrast, for wet etched devices it stayed constant with density. Moreover, the boundary scattering was found to be more specular for wet-etched devices, which may be relevant for studying topological edge states. | Influence of etching processes on electronic transport in mesoscopic InAs/GaSb quantum well devices |
Let 0<p<1 be fixed. Shamir and Spencer proved in the 1980s that the chromatic number of a random graph in G(n,p) is concentrated in an interval of length about n^{1/2}. In this explanatory note, we give a proof of a result due due Noga Alon, showing that the chromatic number is concentrated in an interval of length about n^{1/2}/log n. | On the concentration of the chromatic number of random graphs |
We discuss the single diffractive production of $c \bar c$ pairs and charmed mesons at the LHC. In addition to standard collinear approach, for a first time we propose a $k_t$-factorization approach to the diffractive processes. The transverse momentum dependent (the unintegrated diffractive parton distributions) in proton are obtained with the help of the Kimber-Martin-Ryskin prescription where collinear diffractive PDFs are used as input. In this calculation the transverse momentum of the pomeron is neglected with respect to transverse momentum of partons entering the hard process. The results of the new approach are compared with those of the standard collinear one. Significantly larger cross sections are obtained in the $k_t$-factorization approach where some part of higher-order effects is effectively included. Some correlation observables, like azimuthal angle correlation between $c$ and $\bar c$, and $c \bar c$ pair transverse momentum distribution were obtained for the first time. | Diffractive processes at the LHC within $k_t$ -factorization approach |
We investigate linear and non-linear transport in a double quantum dot system weakly coupled to spin-polarized leads. In the linear regime, the conductance as well as the non-equilibrium spin accumulation are evaluated in analytic form. The conductance as a function of the gate voltage exhibits four peaks of different height, with mirror symmetry with respect to the charge neutrality point. As the polarization angle is varied, the position and shape of the peaks changes in a characteristic way which preserves the electron-hole symmetry of the problem. In the nonlinear regime negative differential conductance features occur for non collinear magnetisations of the leads. In the considered sequential tunneling limit, the tunneling magneto resistance (TMR) is always positive with a characteristic gate voltage dependence for non-collinear magnetization. If a magnetic field is added to the system, the TMR can become negative. | Transport through a double quantum dot system with non-collinearly polarized leads |
We show that gravitating Merons in $D$-dimensional massive Yang-Mills theory can be mapped to solutions of the Einstein-Skyrme model. The identification of the solutions relies on the fact that, when considering the Meron ansatz for the gauge connection $A=\lambda U^{-1}dU$, the massive Yang-Mills equations reduce to the Skyrme equations for the corresponding group element $U$. In the same way, the energy-momentum tensors of both theories can be identified and therefore lead to the same Einstein equations. Subsequently, we focus on the $SU(2)$ case and show that introducing a mass for the Yang-Mills field restricts Merons to live on geometries given by the direct product of $S^3$ (or $S^2$) and Lorentzian manifolds with constant Ricci scalar. We construct explicit examples for $D=4$ and $D=5$. Finally, we comment on possible generalisations. | Gravitating Meron-like topological solitons in massive Yang-Mills theory and the Einstein-Skyrme model |
Systems of $N$ identical globally coupled phase oscillators can demonstrate a multitude of complex behaviours. Such systems can have chaotic dynamics for $N>4$ when a coupling function is biharmonic. The case $N = 4$ does not possess chaotic attractors when the coupling is biharmonic, but has them when the coupling includes nonpairwise interactions of phases. Previous studies showed that some of chaotic attractors in this system are organized by heteroclinic networks. In present paper we discuss which heteroclinic cycles are forbidden and which are supported by this particular system. We also discuss some of the cases regarding homoclinic trajectories to saddle-foci equilibria. | Heteroclinic and homoclinic structures in the system of four identical globally coupled phase oscillators with nonpairwise interactions of phases |
Bohmian mechanics, widely known within the field of the quantum foundations, has been a quite useful resource for computational and interpretive purposes in a wide variety of practical problems. Here, it is used to establish a comparative analysis at different levels of approximation in the problem of the diffraction of helium atoms from a substrate consisting of a defect with axial symmetry on top of a flat surface. The motivation behind this work is to determine which aspects of one level survive in the next level of refinement and, therefore, to get a better idea of what we usually denote as quantum-classical correspondence. To this end, first a quantum treatment of the problem is performed with both an approximated hard-wall model and then with a realistic interaction potential model. The interpretation and explanation of the features displayed by the corresponding diffraction intensity patterns is then revisited with a series of trajectory-based approaches: Fermatian trajectories (optical rays), Newtonian trajectories and Bohmian trajectories. As it is seen, while Fermatian and Newtonian trajectories show some similarities, Bohmian trajectories behave quite differently due to their implicit non-classicality. | Atom-diffraction from surfaces with defects: A Fermatian, Newtonian and Bohmian joint view |
Conditional detection is an important tool to extract weak signals from a noisy background and is closely linked to heralding, which is an essential component of protocols for long distance quantum communication and distributed quantum information processing in quantum networks. Here we demonstrate the conditional detection of time-bin qubits after storage in and retrieval from a photon-echo based waveguide quantum memory. Each qubit is encoded into one member of a photon-pair produced via spontaneous parametric down conversion, and the conditioning is achieved by the detection of the other member of the pair. Performing projection measurements with the stored and retrieved photons onto different bases we obtain an average storage fidelity of 0.885 \pm 0.020, which exceeds the relevant classical bounds and shows the suitability of our integrated light-matter interface for future applications of quantum information processing. | Conditional detection of pure quantum states of light after storage in a waveguide |
We perform the fit of electroweak precision observables within the Standard Model with a 126 GeV Higgs boson, compare the results with the theoretical predictions and discuss the impact of recent experimental and theoretical improvements. We introduce New Physics contributions in a model-independent way and fit for the S, T and U parameters, for the $\epsilon_{1,2,3,b}$ ones, for modified $Zb\bar{b}$ couplings and for a modified Higgs coupling to vector bosons. We point out that composite Higgs models are very strongly constrained. Finally, we compute the bounds on dimension-six operators relevant for the electroweak fit. | Electroweak Precision Observables, New Physics and the Nature of a 126 GeV Higgs Boson |
The thermodynamics of the Kerr-AdS black hole is reformulated within the context of the formalism of geometrothermodynamics (GTD) and the cosmological constant is considered as a new thermodynamical parameter. We conclude that the mass of the black hole corresponds to the total enthalpy of this system. Choosing appropriately the metric in the equilibrium states manifold, we study the phase transitions as a divergence of the thermodynamical curvature scalar. This approach reproduces the Hawking-Page transition and shows that considering the cosmological constant as a thermodynamical parameter does not contribute with new phase transitions. | Geometric Thermodynamics of Kerr-AdS black hole with a Cosmological Constant as State Variable |
Stress is various mental health disorders including depression and anxiety among college students. Early stress diagnosis and intervention may lower the risk of developing mental illnesses. We examined a machine learning-based method for identification of stress using data collected in a naturalistic study utilizing self-reported stress as ground truth as well as physiological data such as heart rate and hand acceleration. The study involved 54 college students from a large campus who used wearable wrist-worn sensors and a mobile health (mHealth) application continuously for 40 days. The app gathered physiological data including heart rate and hand acceleration at one hertz frequency. The application also enabled users to self-report stress by tapping on the watch face, resulting in a time-stamped record of the self-reported stress. We created, evaluated, and analyzed machine learning algorithms for identifying stress episodes among college students using heart rate and accelerometer data. The XGBoost method was the most reliable model with an AUC of 0.64 and an accuracy of 84.5%. The standard deviation of hand acceleration, standard deviation of heart rate, and the minimum heart rate were the most important features for stress detection. This evidence may support the efficacy of identifying patterns in physiological reaction to stress using smartwatch sensors and may inform the design of future tools for real-time detection of stress. | Evaluating Mental Stress Among College Students Using Heart Rate and Hand Acceleration Data Collected from Wearable Sensors |
We tackle the problem of graph out-of-distribution (OOD) generalization. Existing graph OOD algorithms either rely on restricted assumptions or fail to exploit environment information in training data. In this work, we propose to simultaneously incorporate label and environment causal independence (LECI) to fully make use of label and environment information, thereby addressing the challenges faced by prior methods on identifying causal and invariant subgraphs. We further develop an adversarial training strategy to jointly optimize these two properties for causal subgraph discovery with theoretical guarantees. Extensive experiments and analysis show that LECI significantly outperforms prior methods on both synthetic and real-world datasets, establishing LECI as a practical and effective solution for graph OOD generalization. | Joint Learning of Label and Environment Causal Independence for Graph Out-of-Distribution Generalization |
Cross-lingual or cross-domain correspondences play key roles in tasks ranging from machine translation to transfer learning. Recently, purely unsupervised methods operating on monolingual embeddings have become effective alignment tools. Current state-of-the-art methods, however, involve multiple steps, including heuristic post-hoc refinement strategies. In this paper, we cast the correspondence problem directly as an optimal transport (OT) problem, building on the idea that word embeddings arise from metric recovery algorithms. Indeed, we exploit the Gromov-Wasserstein distance that measures how similarities between pairs of words relate across languages. We show that our OT objective can be estimated efficiently, requires little or no tuning, and results in performance comparable with the state-of-the-art in various unsupervised word translation tasks. | Gromov-Wasserstein Alignment of Word Embedding Spaces |
We construct a simple phenomenological diffuse-interface model for composition-induced nanopatterning during ion sputtering of alloys. In simulations, this model reproduces without difficulties the high-aspect ratio structures and tilted pillars observed in experiments. We investigate the time evolution of the pillar height, both by simulations and by {\it in situ} ellipsometry. The analysis of the simulation results yields a good understanding of the transitions between different growth regimes and supports the role of segregation in the pattern-formation process. | Diffuse-interface model for nanopatterning induced by self-sustained ion etch masking |
The regression discontinuity design (RDD) is a quasi-experimental design that can be used to identify and estimate the causal effect of a treatment using observational data. In an RDD, a pre-specified rule is used for treatment assignment, whereby a subject is assigned to the treatment (control) group whenever their observed value of a specific continuous variable is greater than or equal to (is less than) a fixed threshold. Sharp RDDs occur when guidelines are strictly adhered to and fuzzy RDDs occur when the guidelines or not strictly adhered to. In this paper, we take a rigorous decision theoretic approach to formally study causal effect identification and estimation in both sharp and fuzzy RDDs. We use the language and calculus of conditional independence to express and explore in a clear and precise manner the conditions implied by each RDD and investigate additional assumptions under which the identification of the average causal effect at the threshold can be achieved. We apply the methodology in an example concerning the relationship between statins (a class of cholesterol-lowering drugs) and low density lipoprotein (LDL) cholesterol using a real set of primary care data. | Regression Discontinuity Designs: A Decision Theoretic Approach |
London penetration depth, $\lambda(T)$, was measured in single crystals of SrFe$_2$(As$_{1-x}$P$_x$)$_2$ ($x=$0.35) iron - based superconductor. The influence of disorder on the transition temperature, $T_c$, and on $\lambda(T)$ was investigated. The effects of scattering controlled by the annealing of as-grown crystals was compared with the effects of artificial disorder introduced by 2.5~MeV electron irradiation. The low temperature behavior of $\lambda(T)$ can be described by a power-law function, $\Delta \lambda (T)=AT^n$, with the exponent $n$ close to one in pristine annealed samples, as expected for superconducting gap with line nodes. Upon $1.2 \times 10^{19}$ \ecm irradiation, the exponent $n$ increases rapidly exceeding a dirty limit value of $n=$ 2 implying that the nodes in the superconducting gap are accidental and can be lifted by the disorder. The variation of the exponent $n$ with $T_c$ is much stronger in the irradiated crystals compared to the crystals in which disorder was controlled by the annealing of the growth defects. We discuss the results in terms of different influence of different types of disorder on intra- and inter- band scattering. | Comparative study of the effects of electron irradiation and natural disorder in single crystals of SrFe$_{2}$(As$_{1-x}$P$_x$)$_2$ ($x=$0.35) superconductor |
A possible detection of sub-solar mass ultra-compact objects would lead to new perspectives on the existence of black holes that are not of astrophysical origin and/or pertain to formation scenarios of exotic ultra-compact objects. Both possibilities open new perspectives for better understanding of our universe. In this work, we investigate the significance of detection of sub-solar mass binaries with components mass in the range: $10^{-2} M_\odot$ up to 1$M_\odot$, within the expected sensitivity of the ground-based gravitational waves detectors of third-generation, viz., the Einstein Telescope (ET) and the Cosmic Explorer (CE). Assuming a minimum of amplitude signal-to-noise ratio for detection, viz., $\rho = 8$, we find that the maximum horizon distances for an ultra-compact binary system with components mass $10^{-2} \, M_\odot$ and 1$M_\odot$ are 40 Mpc and 1.89 Gpc, respectively, for ET, and 125 Mpc and 5.8 Gpc, respectively, for CE. Other cases are also presented in the text. We derive the merger rate, and discuss consequences on the abundances of primordial black hole (PBH), $f_{\rm PBH}$. Considering the entire mass range [$10^{-2}$ - 1]$M_\odot$, we find $f_{\rm PBH} < 0.70$ ($<$ $0.06$) for ET (CE), respectively. | Search for sub-solar mass binaries with Einstein Telescope and Cosmic Explorer |
In the article the formulas for the modeling of conservative fields in piecewise infinite plate with a thin inclusion found. The accuracy of the found formulas is of order equal to the thickness of the outer layer. The problem for higher accuracy solved by means of asymptotic formulas. | Modeling of conservative fields in piecewise infinite plate with a thin inclusion |
We optimize a collision-induced cooling process for ultracold atoms in the nondegenerate regime. It makes use of a Feshbach resonance, instead of rf radiation in evaporative cooling, to selectively expel hot atoms from a trap. Using functional minimization we analytically show that for the optimal cooling process the resonance energy must be tuned such that it linearly follows the temperature. Here, optimal cooling is defined as maximizing the phase-space density after a fixed cooling duration. The analytical results are confirmed by numerical Monte-Carlo simulations. In order to simulate more realistic experimental conditions, we show that background losses do not change our conclusions, while additional non-resonant two-body losses make a lower initial resonance energy with non-linear dependence on temperature preferable. | Optimization of collisional Feshbach cooling of an ultracold nondegenerate gas |
A sequence of zero-temperature black-hole spacetimes with angular momentum and electric and magnetic charges is shown to exist in gauged $N=2$ supergravity. Stability of a subset of these spacetimes is demonstrated by saturation of the Bogomol'nyi bound arising from the supersymmetry algebra. The mass of the resulting solitonic black holes is given in terms of the cosmological constant and the angular momentum. We conjecture that at the quantum level these solitons are dyons with angular momentum determined by the electric and magnetic charges. | Solitonic Black Holes in Gauged N=2 Supergravity |
Low mass galaxies are expected to be dark matter dominated even within their centrals. Recently two observations reported two dwarf galaxies in group environment with very little dark matter in their centrals. We explore the population and origins of dark-matter-deficient galaxies (DMDGs) in two state-of-the-art hydrodynamical simulations, the EAGLE and Illustris projects. For all satellite galaxies with $10^9<M_*<10^{10}$ M$_{\odot}$ in groups with $M_{200}>10^{13}$ M$_{\odot}$, we find that about $2.6\%$ of them in the EAGLE, and $1.5\%$ in the Illustris are DMDGs with dark matter fractions below $50\%$ inside two times half-stellar-mass radii. We demonstrate that DMDGs are highly tidal disrupted galaxies; and because dark matter has higher binding energy than stars, mass loss of the dark matter is much more rapid than stars in DMDGs during tidal interactions. If DMDGs were confirmed in observations, they are expected in current galaxy formation models. | Dark-matter-deficient galaxies in hydrodynamical simulations |
We consider locally isotropic Gaussian random fields on the $N$-dimensional Euclidean space for fixed $N$. Using the so called Gaussian Orthogonally Invariant matrices first studied by Mallows in 1961 which include the celebrated Gaussian Orthogonal Ensemble (GOE), we establish the Kac--Rice representation of expected number of critical points of non-isotropic Gaussian fields, complementing the isotropic case obatined by Cheng and Schwartzman in 2018. In the limit $N=\8$, we show that such a representation can be always given by GOE matrices, as conjectured by Auffinger and Zeng in 2020. | On the expected number of critical points of locally isotropic Gaussian random fields |
In a recent work, the degenerate Stirling polynomials of the second kind were studied by T. Kim. In this paper, we investigate the extended degenerate Stirling numbers of the second kind and the extended degenerate Bell polynomials associated with them. As results, we give some expressions, identities and properties about the extended degener- ate Stirling numbers of the second kind and the extended degenerate Bell polynomials. | Extended degenerate Stirling numbers of the second kind and extended degenerate Bell polynomials |
In this paper we present new near infrared (NIR) imaging and spectroscopic data of the quasar 3C 48 and its host galaxy. The data were obtained with the ESO-VLT camera ISAAC.We report the first detection of the apparent second nucleus 3C 48A about 100NE of the bright QSO nucleus in the NIR bands J, H, and Ks. 3C 48A is highly reddened with respect to the host, which could be due to warm dust, heated by enhanced star formation or by interstellar material intercepting the radio jet. In fact, all colors on the host galaxy are reddened by several magnitudes of visual extinction. Imaging and initial spectroscopy also reveal a stellar content of about 30% to the overall QSO-light in the NIR. These results are important input parameters for future models of the stellar populations by taking extinction into account. | NIR Observations of the QSO 3C 48 Host Galaxy |
This report studies data-driven estimation of the directed information (DI) measure between two{em discrete-time and continuous-amplitude} random process, based on the $k$-nearest-neighbors ($k$-NN) estimation framework. Detailed derivations of two $k$-NN estimators are provided. The two estimators differ in the metric based on which the nearest-neighbors are found. To facilitate the estimation of the DI measure, it is assumed that the observed sequences are (jointly) Markovian of order $m$. As $m$ is generally not known, a data-driven method (that is also based on the $k$-NN principle) for estimating $m$ from the observed sequences is presented. An exhaustive numerical study shows that the discussed $k$-NN estimators perform well even for relatively small number of samples (few thousands). Moreover, it is shown that the discussed estimators are capable of accurately detecting linear as well as non-linear causal interactions. | $k$-NN Estimation of Directed Information |
This paper presents the design and the realization of a 60 GHz wireless Gigabit Ethernet communication system. A differential encoded binary phase shift keying modulation (DBPSK) and differential demodulation schemes are adopted for the IF blocks. The Gigabit Ethernet interface allows a high speed transfer of multimedia files via a 60 GHz wireless link. First measurement results are shown for 875 Mbps data rate. | 60 GHz High Data Rate Wireless Communication System |
We show the robustness of the structure of Legendre transform in thermodynamics against the replacement of the standard linear average with the Kolmogorov-Nagumo nonlinear average to evaluate the expectation values of the macroscopic physical observables. The consequence of this statement is twofold: 1) the relationships between the expectation values and the corresponding Lagrange multipliers still hold in the present formalism; 2) the universality of the Gibbs equation as well as other thermodynamic relations are unaffected by the structure of the average used in the theory. | Consistency of the structure of Legendre transform in thermodynamics with the Kolmogorov-Nagumo average |
A complete representation of the Martin boundary of killed random walks on the quadrant ${\mathbb{N}}^*\times{\mathbb{N}}^*$ is obtained. It is proved that the corresponding full Martin compactification of the quadrant ${\mathbb{N}}^*\times{\mathbb{N}}^*$ is homeomorphic to the closure of the set $\{w={z}/{(1+|z|)}:z\in{\mathbb{N}}^*\times{\mathbb{N}}^*\}$ in ${\mathbb{R}}^2$. The method is based on a ratio limit theorem for local processes and large deviation techniques. | Martin boundary of a killed random walk on a quadrant |
We investigate the $L^2$-supercritical and $\dot{H}^1$-subcritical nonlinear Schr\"{o}dinger equation in $H^1$. In \cite{G1} and \cite{yuan}, the mass-energy quantity $M(Q)^{\frac{1-s_{c}}{s_{c}}}E(Q)$ has been shown to be a threshold for the dynamical behavior of solutions of the equation. In the present paper, we study the dynamics at the critical level $M(u)^{\frac{1-s_{c}}{s_{c}}}E(u)=M(Q)^{\frac{1-s_{c}}{s_{c}}}E(Q)$ and classify the corresponding solutions using modulation theory, non-trivially generalize the results obtained in \cite{holmer3} for the 3D cubic Schr\"{o}dinger equation. | Threshold solutions for the focusing $L^{2}$ -supercritical NLS Equations |
We present a new software system PETSc TSAdjoint for first-order and second-order adjoint sensitivity analysis of time-dependent nonlinear differential equations. The derivative calculation in PETSc TSAdjoint is essentially a high-level algorithmic differentiation process. The adjoint models are derived by differentiating the timestepping algorithms and implemented based on the parallel infrastructure in PETSc. Full differentiation of the library code including MPI routines thus is avoided, and users do not need to derive their own adjoint models for their specific applications. PETSc TSAdjoint can compute the first-order derivative, that is, the gradient of a scalar functional, and the Hessian-vector product that carries second-order derivative information, while requiring minimal input (a few callbacks) from the users. Optimal checkpointing schemes are employed by the adjoint model in a manner that is transparent to users. Usability, efficiency, and scalability are demonstrated through examples from a variety of applications. | PETSc TSAdjoint: a discrete adjoint ODE solver for first-order and second-order sensitivity analysis |
In this paper, we study numerical homogenization methods based on integral equations. Our work is motivated by materials such as concrete, modeled as composites structured as randomly distributed inclusions imbedded in a matrix. We investigate two integral reformulations of the corrector problem to be solved, namely the equivalent inclusion method based on the Lippmann-Schwinger equation, and a method based on boundary integral equations. The fully populated matrices obtained by the discretization of the integral operators are successfully dealt with using the H-matrix format. | Application of Hierarchical Matrix Techniques To The Homogenization of Composite Materials |
Understanding the ultrafast carrier dynamics of graphene on a substrate is a fundamental step in the development of graphene based opto-electronic devices. Here, we present ultrafast pump-probe measurements of supported graphene on quartz for a range of pump fluences that enable us to observe both decreased and enhanced probe transmission regimes on a femtosecond timescale. Unexpectedly, at an intermediate pump fluence, an order of magnitude decrease in the relaxation time constant of the differential transmission is observed. By employing a number of different models to interpret our experimental data, we demonstrate the importance of considering both intra- and inter-band contributions to the dynamical optical conductivity in order to extract a more physical relaxation time constant of hot optical phonons. | On the interpretation of manifold ultrafast dynamics in supported graphene |
We derive and analyze the alternating direction explicit (ADE) method for time evolution equations with the time-dependent Dirichlet boundary condition and with the zero Neumann boundary condition. The original ADE method is an additive operator splitting (AOS) method, which has been developed for treating a wide range of linear and nonlinear time evolution equations with the zero Dirichlet boundary condition. For linear equations, it has been shown to achieve the second order accuracy in time yet is unconditionally stable for an arbitrary time step size. For the boundary conditions considered in this work, we carefully construct the updating formula at grid points near the boundary of the computational domain and show that these formulas maintain the desired accuracy and the property of unconditional stability. We also construct numerical methods based on the ADE scheme for two classes of fractional differential equations. We will give numerical examples to demonstrate the simplicity and the computational efficiency of the method. | An Alternating Direction Explicit Method for Time Evolution Equations with Applications to Fractional Differential Equations |
Deep neural networks have emerged as effective tools for computational imaging including quantitative phase microscopy of transparent samples. To reconstruct phase from intensity, current approaches rely on supervised learning with training examples; consequently, their performance is sensitive to a match of training and imaging settings. Here we propose a new approach to phase microscopy by using an untrained deep neural network for measurement formation, encapsulating the image prior and imaging physics. Our approach does not require any training data and simultaneously reconstructs the sought phase and pupil-plane aberrations by fitting the weights of the network to the captured images. To demonstrate experimentally, we reconstruct quantitative phase from through-focus images blindly (i.e. no explicit knowledge of the aberrations). | Deep Phase Decoder: Self-calibrating phase microscopy with an untrained deep neural network |
We construct the Baxter operator $\boldsymbol{ \texttt{Q} }(\lambda)$ for the $q$-Toda chain and the Toda$_2$ chain (the Toda chain in the second Hamiltonian structure). Our construction builds on the relation between the Baxter operator and B\"acklund transformations that were unravelled in {\cite{GaPa92}}. We construct a number of quantum intertwiners ensuring the commutativity of $\boldsymbol{ \texttt{Q} }(\lambda)$ with the transfer matrix of the models and the one of $\boldsymbol{ \texttt{Q} }$'s between each other. Most importantly, $\boldsymbol{ \texttt{Q} }(\lambda)$ is modular invariant in the sense of Faddeev. We derive the Baxter equation for the eigenvalues $q(\lambda)$ of $\boldsymbol{ \texttt{Q} }(\lambda)$ and show that these are entire functions of $\lambda$. This last property will ultimately lead to the quantisation of the spectrum for the considered Toda chains, in a subsequent publication. | Baxter operator and Baxter equation for $q$-Toda and Toda$_2$ chains |
In superconductors, and in other systems with a local U(1) gauge invariance, there are two mechanisms that form topological defects in phase transitions. In addition to the standard Kibble mechanism, thermal fluctuations of the magnetic field also lead to defect formation. This mechanism is specific to local gauge theories, predicts a qualitatively different spatial defect distribution and is the dominant source for topological defects in slow transitions. I review the arguments that lead to these conclusions and discuss possibilities of testing the scenario in superconductor experiments. | Local gauge invariance and formation of topological defects |
Minimum flow decomposition (MFD) is the NP-hard problem of finding a smallest decomposition of a network flow/circulation $X$ on a directed graph $G$ into weighted source-to-sink paths whose superposition equals $X$. We show that, for acyclic graphs, considering the \emph{width} of the graph (the minimum number of paths needed to cover all of its edges) yields advances in our understanding of its approximability. For the version of the problem that uses only non-negative weights, we identify and characterise a new class of \emph{width-stable} graphs, for which a popular heuristic is a \gwsimple-approximation ($|X|$ being the total flow of $X$), and strengthen its worst-case approximation ratio from $\Omega(\sqrt{m})$ to $\Omega(m / \log m)$ for sparse graphs, where $m$ is the number of edges in the graph. We also study a new problem on graphs with cycles, Minimum Cost Circulation Decomposition (MCCD), and show that it generalises MFD through a simple reduction. For the version allowing also negative weights, we give a $(\lceil \log \Vert X \Vert \rceil +1)$-approximation ($\Vert X \Vert$ being the maximum absolute value of $X$ on any edge) using a power-of-two approach, combined with parity fixing arguments and a decomposition of unitary circulations ($\Vert X \Vert \leq 1$), using a generalised notion of width for this problem. Finally, we disprove a conjecture about the linear independence of minimum (non-negative) flow decompositions posed by Kloster et al. [ALENEX 2018], but show that its useful implication (polynomial-time assignments of weights to a given set of paths to decompose a flow) holds for the negative version. | Width Helps and Hinders Splitting Flows |
We present a new, probabilistic method for determining the systemic proper motions of Milky Way (MW) ultra-faint satellites in the Dark Energy Survey (DES). We utilize the superb photometry from the first public data release (DR1) of DES to select candidate members, and cross-match them with the proper motions from $Gaia$ DR2. We model the candidate members with a mixture model (satellite and MW) in spatial and proper motion space. This method does not require prior knowledge of satellite membership, and can successfully determine the tangential motion of thirteen DES satellites. With our method we present measurements of the following satellites: Columba~I, Eridanus~III, Grus~II, Phoenix~II, Pictor~I, Reticulum~III, and Tucana~IV; this is the first systemic proper motion measurement for several and the majority lack extensive spectroscopic follow-up studies. We compare these to the predictions of Large Magellanic Cloud satellites and to the vast polar structure. With the high precision DES photometry we conclude that most of the newly identified member stars are very metal-poor ([Fe/H] $\lesssim -2$) similar to other ultra-faint dwarf galaxies, while Reticulum III is likely more metal-rich. We also find potential members in the following satellites that might indicate their overall proper motion: Cetus~II, Kim~2, and Horologium~II; however, due to the small number of members in each satellite, spectroscopic follow-up observations are necessary to determine the systemic proper motion in these satellites. | Proper motions of Milky Way Ultra-Faint satellites with ${\it Gaia}$ DR2 $\times$ DES DR1 |
Time series both of microwave radiometer brightness temperature measurements at 23.8 and 31.4 GHz and of retrievals of water vapor and liquid water path from these brightness temperatures are evaluated using the detrended fluctuation analysis method. As quantified by the parameter $\alpha$, this method (i) enables identification of the time scales over which noise dominates the time series and (ii) characterizes the temporal range of correlations in the time series. The more common spectral analysis method is also used to assess the data and its results are compared with those from detrended fluctuation analysis method. The assumption that measurements should have certain scaling properties allows the quality of the measurements to be characterized. The additional assumption that the scaling properties of the measurements of an atmospheric quantity are preserved in a useful retrieval provides a means for evaluating the retrieval itself. Applying these two assumptions to microwave radiometer measurements and retrievals demonstrates three points. First, the retrieved water vapor path during cloudy-sky periods can be dominated by noise on shorter than ~30~min time scales ($\alpha$-exponent = 0.1) and exhibits no scaling behavior at longer time scales. However, correlations in the brightness temperatures and liquid water path retrievals are found to be consistent with a power-law behavior for time scales up to 3 hr with an $\alpha$-exponent equal to approximately 0.3, as in other geophysical phenomena. Second, clear-sky, moist atmospheres show the expected scaling for both measurements and retrievals of the water vapor path. Third, during clear-sky, dry atmospheric days, instrument noise from the 31.4 GHz channel compromises the quality of the water vapor path retrieval. | Evaluating the quality of ground-based microwave radiometer measurements and retrievals using detrended fluctuation and spectral analysis methods |
We discuss the criteria to associate PSRs with SNRs, and summary 78 pairs of possible PSR-SNR associations which is the most complete sample so far. We refine them into three categories according to degree of reliability. Statistic study on PSR-SNR associations helps us understand massive star evolution and constrain pulsar's theory models. | 78 Pairs of Possible PSR-SNR Associations |
In this paper, we show that the global monopole spacetime is one of the exact solutions of Einstein equations by means of the method treating the matter field as a non-linear sigma model, without the weak field approximation applied in the original derivation by Barriola and Vilenkin in 1989. Furthermore, we find the physical origin of the topological charge in the global monopole spacetime. Finally, we generalize the proposal which generates spacetime from thermodynamical laws to the case that spacetime with global monopole charge. | The global monopole spacetime and its topological charge |
Value function approximation is a crucial module for policy evaluation in reinforcement learning when the state space is large or continuous. The present paper takes a generative perspective on policy evaluation via temporal-difference (TD) learning, where a Gaussian process (GP) prior is presumed on the sought value function, and instantaneous rewards are probabilistically generated based on value function evaluations at two consecutive states. Capitalizing on a random feature-based approximant of the GP prior, an online scalable (OS) approach, termed {OS-GPTD}, is developed to estimate the value function for a given policy by observing a sequence of state-reward pairs. To benchmark the performance of OS-GPTD even in an adversarial setting, where the modeling assumptions are violated, complementary worst-case analyses are performed by upper-bounding the cumulative Bellman error as well as the long-term reward prediction error, relative to their counterparts from a fixed value function estimator with the entire state-reward trajectory in hindsight. Moreover, to alleviate the limited expressiveness associated with a single fixed kernel, a weighted ensemble (E) of GP priors is employed to yield an alternative scheme, termed OS-EGPTD, that can jointly infer the value function, and select interactively the EGP kernel on-the-fly. Finally, performances of the novel OS-(E)GPTD schemes are evaluated on two benchmark problems. | Robust and Adaptive Temporal-Difference Learning Using An Ensemble of Gaussian Processes |
MPI has been ubiquitously deployed in flagship HPC systems aiming to accelerate distributed scientific applications running on tens of hundreds of processes and compute nodes. Maintaining the correctness and integrity of MPI application execution is critical, especially for safety-critical scientific applications. Therefore, a collection of effective MPI fault tolerance techniques have been proposed to enable MPI application execution to efficiently resume from system failures. However, there is no structured way to study and compare different MPI fault tolerance designs, so to guide the selection and development of efficient MPI fault tolerance techniques for distinct scenarios. To solve this problem, we design, develop, and evaluate a benchmark suite called MATCH to characterize, research, and comprehensively compare different combinations and configurations of MPI fault tolerance designs. Our investigation derives useful findings: (1) Reinit recovery in general performs better than ULFM recovery; (2) Reinit recovery is independent of the scaling size and the input problem size, whereas ULFM recovery is not; (3) Using Reinit recovery with FTI checkpointing is a highly efficient fault tolerance design. MATCH code is available at https://github.com/kakulo/MPI- FT- Bench. | MATCH: An MPI Fault Tolerance Benchmark Suite |
High energy gamma-ray emission from two nearby bright starburst galaxies, M82 and NGC 253, have recently been detected by Fermi, H.E.S.S., and VERITAS. Since starburst galaxies have a high star formation rate and plenty of dust in the central starburst region, infrared emissions are strong there. Gamma-ray photons are absorbed by the interstellar radiation field photons via electron and positron pair creation. The generated electron and positron pairs up scatter the interstellar photons to very high energy gamma-ray photons via cascade emission through inverse Compton scattering. In this paper, we evaluate the contribution of this cascade emission to the gamma-ray spectra of M82 and NGC 253. Although it would be difficult to see direct gamma- ray evidence of cosmic-rays with an energy > 10 TeV due to the gamma-ray attenuation, the resulting cascade emission would be indirect evidence. By including the cascade component, we find that the total flux above 1 TeV increases ~18% and ~45% compared with the absorbed flux assuming the maximum kinetic proton energy as 45.3 TeV and 512 TeV, respectively. Future gamma-ray observatories such as CTA would be able to see the indirect evidence of cosmic-ray with an energy > 10 TeV by comparing with theoretical emission models including this cascade effect. | High Energy Gamma-ray Absorption and Cascade Emission in Nearby Starburst Galaxies |
For every positive regular Borel measure, possibly infinite valued, vanishing on all sets of $p$-capacity zero, we characterize the compactness of the embedding $W^{1,p}({\bf R}^N)\cap L^p ({\bf R}^N,\mu)\hr L^q({\bf R}^N)$ in terms of the qualitative behavior of some characteristic PDE. This question is related to the well posedness of a class of geometric inequalities involving the torsional rigidity and the spectrum of the Dirichlet Laplacian introduced by Polya and Szeg\"o in 1951. In particular, we prove that finite torsional rigidity of an arbitrary domain (possibly with infinite measure), implies the compactness of the resolvent of the Laplacian. | On the characterization of the compact embedding of Sobolev spaces |
Using constructions of the Whitham perturbation theory of integrable system we prove a new sharp upper bound of $3g/2-2$ on the dimension of complete subvarieties of $\M_g^{ct}$. | Real normalized differentials and compact cycles in the moduli space of curves |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.