title
stringlengths 7
239
| abstract
stringlengths 7
2.76k
| cs
int64 0
1
| phy
int64 0
1
| math
int64 0
1
| stat
int64 0
1
| quantitative biology
int64 0
1
| quantitative finance
int64 0
1
|
|---|---|---|---|---|---|---|---|
On the Statistical Efficiency of Optimal Kernel Sum Classifiers | We propose a novel combination of optimization tools with learning theory
bounds in order to analyze the sample complexity of optimal kernel sum
classifiers. This contrasts the typical learning theoretic results which hold
for all (potentially suboptimal) classifiers. Our work also justifies
assumptions made in prior work on multiple kernel learning. As a byproduct of
our analysis, we also provide a new form of Rademacher complexity for
hypothesis classes containing only optimal classifiers.
| 1 | 0 | 0 | 1 | 0 | 0 |
Ultracold atoms in multiple-radiofrequency dressed adiabatic potentials | We present the first experimental demonstration of a multiple-radiofrequency
dressed potential for the configurable magnetic confinement of ultracold atoms.
We load cold $^{87}$Rb atoms into a double well potential with an adjustable
barrier height, formed by three radiofrequencies applied to atoms in a static
quadrupole magnetic field. Our multiple-radiofrequency approach gives precise
control over the double well characteristics, including the depth of individual
wells and the height of the barrier, and enables reliable transfer of atoms
between the available trapping geometries. We have characterised the
multiple-radiofrequency dressed system using radiofrequency spectroscopy,
finding good agreement with the eigenvalues numerically calculated using
Floquet theory. This method creates trapping potentials that can be
reconfigured by changing the amplitudes, polarizations and frequencies of the
applied dressing fields, and easily extended with additional dressing
frequencies.
| 0 | 1 | 0 | 0 | 0 | 0 |
Distribution Matching in Variational Inference | We show that Variational Autoencoders consistently fail to learn marginal
distributions in latent and visible space. We ask whether this is a consequence
of matching conditional distributions, or a limitation of explicit model and
posterior distributions. We explore alternatives provided by marginal
distribution matching and implicit distributions through the use of Generative
Adversarial Networks in variational inference. We perform a large-scale
evaluation of several VAE-GAN hybrids and explore the implications of class
probability estimation for learning distributions. We conclude that at present
VAE-GAN hybrids have limited applicability: they are harder to scale, evaluate,
and use for inference compared to VAEs; and they do not improve over the
generation quality of GANs.
| 0 | 0 | 0 | 1 | 0 | 0 |
Henkin constructions of models with size continuum | We survey the technique of constructing customized models of size continuum
in omega steps and illustrate the method by giving new proofs of mostly old
results within this rubric. One new theorem, which is joint with Saharon
Shelah, is that a pseudominimal theory has an atomic model of size continuum.
| 0 | 0 | 1 | 0 | 0 | 0 |
Orthogonal groups in characteristic 2 acting on polytopes of high rank | We show that for all integers $m\geq 2$, and all integers $k\geq 2$, the
orthogonal groups $\Orth^{\pm}(2m,\Fk)$ act on abstract regular polytopes of
rank $2m$, and the symplectic groups $\Sp(2m,\Fk)$ act on abstract regular
polytopes of rank $2m+1$.
| 0 | 0 | 1 | 0 | 0 | 0 |
Free LSD: Prior-Free Visual Landing Site Detection for Autonomous Planes | Full autonomy for fixed-wing unmanned aerial vehicles (UAVs) requires the
capability to autonomously detect potential landing sites in unknown and
unstructured terrain, allowing for self-governed mission completion or handling
of emergency situations. In this work, we propose a perception system
addressing this challenge by detecting landing sites based on their texture and
geometric shape without using any prior knowledge about the environment. The
proposed method considers hazards within the landing region such as terrain
roughness and slope, surrounding obstacles that obscure the landing approach
path, and the local wind field that is estimated by the on-board EKF. The
latter enables applicability of the proposed method on small-scale autonomous
planes without landing gear. A safe approach path is computed based on the UAV
dynamics, expected state estimation and actuator uncertainty, and the on-board
computed elevation map. The proposed framework has been successfully tested on
photo-realistic synthetic datasets and in challenging real-world environments.
| 1 | 0 | 0 | 0 | 0 | 0 |
Semi-supervised and Active-learning Scenarios: Efficient Acoustic Model Refinement for a Low Resource Indian Language | We address the problem of efficient acoustic-model refinement (continuous
retraining) using semi-supervised and active learning for a low resource Indian
language, wherein the low resource constraints are having i) a small labeled
corpus from which to train a baseline `seed' acoustic model and ii) a large
training corpus without orthographic labeling or from which to perform a data
selection for manual labeling at low costs. The proposed semi-supervised
learning decodes the unlabeled large training corpus using the seed model and
through various protocols, selects the decoded utterances with high reliability
using confidence levels (that correlate to the WER of the decoded utterances)
and iterative bootstrapping. The proposed active learning protocol uses
confidence level based metric to select the decoded utterances from the large
unlabeled corpus for further labeling. The semi-supervised learning protocols
can offer a WER reduction, from a poorly trained seed model, by as much as 50%
of the best WER-reduction realizable from the seed model's WER, if the large
corpus were labeled and used for acoustic-model training. The active learning
protocols allow that only 60% of the entire training corpus be manually
labeled, to reach the same performance as the entire data.
| 1 | 0 | 0 | 0 | 0 | 0 |
Energy spectrum of cascade showers generated by cosmic ray muons in water | The spatial distribution of Cherenkov radiation from cascade showers
generated by muons in water has been measured with Cherenkov water calorimeter
(CWC) NEVOD. This result allowed to improve the techniques of treating cascade
showers with unknown axes by means of CWC response analysis. The techniques of
selecting the events with high energy cascade showers and reconstructing their
parameters are discussed. Preliminary results of measurements of the spectrum
of cascade showers in the energy range 100 GeV - 20 TeV generated by cosmic ray
muons at large zenith angles and their comparison with expectation are
presented.
| 0 | 1 | 0 | 0 | 0 | 0 |
The limit point of the pentagram map | The pentagram map is a discrete dynamical system defined on the space of
polygons in the plane. In the first paper on the subject, R. Schwartz proved
that the pentagram map produces from each convex polygon a sequence of
successively smaller polygons that converges exponentially to a point. We
investigate the limit point itself, giving an explicit description of its
Cartesian coordinates as roots of certain degree three polynomials.
| 0 | 0 | 1 | 0 | 0 | 0 |
Reconstruction formulas for Photoacoustic Imaging in Attenuating Media | In this paper we study the problem of photoacoustic inversion in a weakly
attenuating medium. We present explicit reconstruction formulas in such media
and show that the inversion based on such formulas is moderately ill--posed.
Moreover, we present a numerical algorithm for imaging and demonstrate in
numerical experiments the feasibility of this approach.
| 0 | 0 | 1 | 0 | 0 | 0 |
Rank Determination for Low-Rank Data Completion | Recently, fundamental conditions on the sampling patterns have been obtained
for finite completability of low-rank matrices or tensors given the
corresponding ranks. In this paper, we consider the scenario where the rank is
not given and we aim to approximate the unknown rank based on the location of
sampled entries and some given completion. We consider a number of data models,
including single-view matrix, multi-view matrix, CP tensor, tensor-train tensor
and Tucker tensor. For each of these data models, we provide an upper bound on
the rank when an arbitrary low-rank completion is given. We characterize these
bounds both deterministically, i.e., with probability one given that the
sampling pattern satisfies certain combinatorial properties, and
probabilistically, i.e., with high probability given that the sampling
probability is above some threshold. Moreover, for both single-view matrix and
CP tensor, we are able to show that the obtained upper bound is exactly equal
to the unknown rank if the lowest-rank completion is given. Furthermore, we
provide numerical experiments for the case of single-view matrix, where we use
nuclear norm minimization to find a low-rank completion of the sampled data and
we observe that in most of the cases the proposed upper bound on the rank is
equal to the true rank.
| 1 | 0 | 0 | 1 | 0 | 0 |
Network structure from rich but noisy data | Driven by growing interest in the sciences, industry, and among the broader
public, a large number of empirical studies have been conducted in recent years
of the structure of networks ranging from the internet and the world wide web
to biological networks and social networks. The data produced by these
experiments are often rich and multimodal, yet at the same time they may
contain substantial measurement error. In practice, this means that the true
network structure can differ greatly from naive estimates made from the raw
data, and hence that conclusions drawn from those naive estimates may be
significantly in error. In this paper we describe a technique that circumvents
this problem and allows us to make optimal estimates of the true structure of
networks in the presence of both richly textured data and significant
measurement uncertainty. We give example applications to two different social
networks, one derived from face-to-face interactions and one from self-reported
friendships.
| 1 | 1 | 0 | 0 | 0 | 0 |
Algebraic Foundations of Proof Refinement | We contribute a general apparatus for dependent tactic-based proof refinement
in the LCF tradition, in which the statements of subgoals may express a
dependency on the proofs of other subgoals; this form of dependency is
extremely useful and can serve as an algorithmic alternative to extensions of
LCF based on non-local instantiation of schematic variables. Additionally, we
introduce a novel behavioral distinction between refinement rules and tactics
based on naturality. Our framework, called Dependent LCF, is already deployed
in the nascent RedPRL proof assistant for computational cubical type theory.
| 1 | 0 | 0 | 0 | 0 | 0 |
Transfer Learning for Neural Semantic Parsing | The goal of semantic parsing is to map natural language to a machine
interpretable meaning representation language (MRL). One of the constraints
that limits full exploration of deep learning technologies for semantic parsing
is the lack of sufficient annotation training data. In this paper, we propose
using sequence-to-sequence in a multi-task setup for semantic parsing with a
focus on transfer learning. We explore three multi-task architectures for
sequence-to-sequence modeling and compare their performance with an
independently trained model. Our experiments show that the multi-task setup
aids transfer learning from an auxiliary task with large labeled data to a
target task with smaller labeled data. We see absolute accuracy gains ranging
from 1.0% to 4.4% in our in- house data set, and we also see good gains ranging
from 2.5% to 7.0% on the ATIS semantic parsing tasks with syntactic and
semantic auxiliary tasks.
| 1 | 0 | 0 | 0 | 0 | 0 |
Definable Valuations induced by multiplicative subgroups and NIP Fields | We study the algebraic implications of the non-independence property (NIP)
and variants thereof (dp-minimality) on infinite fields, motivated by the
conjecture that all such fields which are neither real closed nor separably
closed admit a definable henselian valuation. Our results mainly focus on Hahn
fields and build up on Will Johnson's preprint "dp-minimal fields", arXiv:
1507.02745v1, July 2015.
| 0 | 0 | 1 | 0 | 0 | 0 |
Bridging the Gap Between Layout Pattern Sampling and Hotspot Detection via Batch Active Learning | Layout hotpot detection is one of the main steps in modern VLSI design. A
typical hotspot detection flow is extremely time consuming due to the
computationally expensive mask optimization and lithographic simulation. Recent
researches try to facilitate the procedure with a reduced flow including
feature extraction, training set generation and hotspot detection, where
feature extraction methods and hotspot detection engines are deeply studied.
However, the performance of hotspot detectors relies highly on the quality of
reference layout libraries which are costly to obtain and usually predetermined
or randomly sampled in previous works. In this paper, we propose an active
learning-based layout pattern sampling and hotspot detection flow, which
simultaneously optimizes the machine learning model and the training set that
aims to achieve similar or better hotspot detection performance with much
smaller number of training instances. Experimental results show that our
proposed method can significantly reduce lithography simulation overhead while
attaining satisfactory detection accuracy on designs under both DUV and EUV
lithography technologies.
| 0 | 0 | 0 | 1 | 0 | 0 |
iCorr : Complex correlation method to detect origin of replication in prokaryotic and eukaryotic genomes | Computational prediction of origin of replication (ORI) has been of great
interest in bioinformatics and several methods including GC Skew, Z curve,
auto-correlation etc. have been explored in the past. In this paper, we have
extended the auto-correlation method to predict ORI location with much higher
resolution for prokaryotes. The proposed complex correlation method (iCorr)
converts the genome sequence into a sequence of complex numbers by mapping the
nucleotides to {+1,-1,+i,-i} instead of {+1,-1} used in the auto-correlation
method (here, 'i' is square root of -1). Thus, the iCorr method uses
information about the positions of all the four nucleotides unlike the earlier
auto-correlation method which uses the positional information of only one
nucleotide. Also, this earlier method required visual inspection of the
obtained graphs to identify the location of origin of replication. The proposed
iCorr method does away with this need and is able to identify the origin
location simply by picking the peak in the iCorr graph. The iCorr method also
works for a much smaller segment size compared to the earlier auto-correlation
method, which can be very helpful in experimental validation of the
computational predictions. We have also developed a variant of the iCorr method
to predict ORI location in eukaryotes and have tested it with the
experimentally known origin locations of S. cerevisiae with an average accuracy
of 71.76%.
| 0 | 1 | 0 | 0 | 0 | 0 |
On the Power Spectral Density Applied to the Analysis of Old Canvases | A routine task for art historians is painting diagnostics, such as dating or
attribution. Signal processing of the X-ray image of a canvas provides useful
information about its fabric. However, previous methods may fail when very old
and deteriorated artworks or simply canvases of small size are studied. We
present a new framework to analyze and further characterize the paintings from
their radiographs. First, we start from a general analysis of lattices and
provide new unifying results about the theoretical spectra of weaves. Then, we
use these results to infer the main structure of the fabric, like the type of
weave and the thread densities. We propose a practical estimation of these
theoretical results from paintings with the averaged power spectral density
(PSD), which provides a more robust tool. Furthermore, we found that the PSD
provides a fingerprint that characterizes the whole canvas. We search and
discuss some distinctive features we may find in that fingerprint. We apply
these results to several masterpieces of the 17th and 18th centuries from the
Museo Nacional del Prado to show that this approach yields accurate results in
thread counting and is very useful for paintings comparison, even in situations
where previous methods fail.
| 1 | 0 | 1 | 0 | 0 | 0 |
Simons' type formula for slant submanifolds of complex space form | In this paper, we study a slant submanifold of a complex space form. We also
obtain an integral formula of Simons' type for a Kaehlerian slant submanifold
in a complex space form and apply it to prove our main result.
| 0 | 0 | 1 | 0 | 0 | 0 |
Eco-evolutionary feedbacks - theoretical models and perspectives | 1. Theoretical models pertaining to feedbacks between ecological and
evolutionary processes are prevalent in multiple biological fields. An
integrative overview is currently lacking, due to little crosstalk between the
fields and the use of different methodological approaches.
2. Here we review a wide range of models of eco-evolutionary feedbacks and
highlight their underlying assumptions. We discuss models where feedbacks occur
both within and between hierarchical levels of ecosystems, including
populations, communities, and abiotic environments, and consider feedbacks
across spatial scales.
3. Identifying the commonalities among feedback models, and the underlying
assumptions, helps us better understand the mechanistic basis of
eco-evolutionary feedbacks. Eco-evolutionary feedbacks can be readily modelled
by coupling demographic and evolutionary formalisms. We provide an overview of
these approaches and suggest future integrative modelling avenues.
4. Our overview highlights that eco-evolutionary feedbacks have been
incorporated in theoretical work for nearly a century. Yet, this work does not
always include the notion of rapid evolution or concurrent ecological and
evolutionary time scales. We discuss the importance of density- and
frequency-dependent selection for feedbacks, as well as the importance of
dispersal as a central linking trait between ecology and evolution in a spatial
context.
| 0 | 0 | 0 | 0 | 1 | 0 |
Unusual behavior of cuprates explained by heterogeneous charge localization | The cuprate high-temperature superconductors are among the most intensively
studied materials, yet essential questions regarding their principal phases and
the transitions between them remain unanswered. Generally thought of as doped
charge-transfer insulators, these complex lamellar oxides exhibit pseudogap,
strange-metal, superconducting and Fermi-liquid behaviour with increasing
hole-dopant concentration. Here we propose a simple inhomogeneous Mott-like
(de)localization model wherein exactly one hole per copper-oxygen unit is
gradually delocalized with increasing doping and temperature. The model is
percolative in nature, with parameters that are experimentally constrained. It
comprehensively captures pivotal unconventional experimental results, including
the temperature and doping dependence of the pseudogap phenomenon, the
strange-metal linear temperature dependence of the planar resistivity, and the
doping dependence of the superfluid density. The success and simplicity of our
model greatly demystify the cuprate phase diagram and point to a local
superconducting pairing mechanism involving the (de)localized hole.
| 0 | 1 | 0 | 0 | 0 | 0 |
An Agile Software Engineering Method to Design Blockchain Applications | Cryptocurrencies and their foundation technology, the Blockchain, are
reshaping finance and economics, allowing a decentralized approach enabling
trusted applications with no trusted counterpart. More recently, the Blockchain
and the programs running on it, called Smart Contracts, are also finding more
and more applications in all fields requiring trust and sound certifications.
Some people have come to the point of saying that the "Blockchain revolution"
can be compared to that of the Internet and the Web in their early days. As a
result, all the software development revolving around the Blockchain technology
is growing at a staggering rate. The feeling of many software engineers about
such huge interest in Blockchain technologies is that of unruled and hurried
software development, a sort of competition on a first-come-first-served basis
which does not assure neither software quality, nor that the basic concepts of
software engineering are taken into account. This paper tries to cope with this
issue, proposing a software development process to gather the requirement,
analyze, design, develop, test and deploy Blockchain applications. The process
is based on several Agile practices, such as User Stories and iterative and
incremental development based on them. However, it makes also use of more
formal notations, such as some UML diagrams describing the design of the
system, with additions to represent specific concepts found in Blockchain
development. The method is described in good detail, and an example is given to
show how it works.
| 1 | 0 | 0 | 0 | 0 | 0 |
Optimizing Prediction Intervals by Tuning Random Forest via Meta-Validation | Recent studies have shown that tuning prediction models increases prediction
accuracy and that Random Forest can be used to construct prediction intervals.
However, to our best knowledge, no study has investigated the need to, and the
manner in which one can, tune Random Forest for optimizing prediction intervals
{ this paper aims to fill this gap. We explore a tuning approach that combines
an effectively exhaustive search with a validation technique on a single Random
Forest parameter. This paper investigates which, out of eight validation
techniques, are beneficial for tuning, i.e., which automatically choose a
Random Forest configuration constructing prediction intervals that are reliable
and with a smaller width than the default configuration. Additionally, we
present and validate three meta-validation techniques to determine which are
beneficial, i.e., those which automatically chose a beneficial validation
technique. This study uses data from our industrial partner (Keymind Inc.) and
the Tukutuku Research Project, related to post-release defect prediction and
Web application effort estimation, respectively. Results from our study
indicate that: i) the default configuration is frequently unreliable, ii) most
of the validation techniques, including previously successfully adopted ones
such as 50/50 holdout and bootstrap, are counterproductive in most of the
cases, and iii) the 75/25 holdout meta-validation technique is always
beneficial; i.e., it avoids the likely counterproductive effects of validation
techniques.
| 0 | 0 | 0 | 1 | 0 | 0 |
Explicit construction of RIP matrices is Ramsey-hard | Matrices $\Phi\in\R^{n\times p}$ satisfying the Restricted Isometry Property
(RIP) are an important ingredient of the compressive sensing methods. While it
is known that random matrices satisfy the RIP with high probability even for
$n=\log^{O(1)}p$, the explicit construction of such matrices defied the
repeated efforts, and the most known approaches hit the so-called $\sqrt{n}$
sparsity bottleneck. The notable exception is the work by Bourgain et al
\cite{bourgain2011explicit} constructing an $n\times p$ RIP matrix with
sparsity $s=\Theta(n^{{1\over 2}+\epsilon})$, but in the regime
$n=\Omega(p^{1-\delta})$.
In this short note we resolve this open question in a sense by showing that
an explicit construction of a matrix satisfying the RIP in the regime
$n=O(\log^2 p)$ and $s=\Theta(n^{1\over 2})$ implies an explicit construction
of a three-colored Ramsey graph on $p$ nodes with clique sizes bounded by
$O(\log^2 p)$ -- a question in the extremal combinatorics which has been open
for decades.
| 0 | 0 | 0 | 1 | 0 | 0 |
Rocket Launching: A Universal and Efficient Framework for Training Well-performing Light Net | Models applied on real time response task, like click-through rate (CTR)
prediction model, require high accuracy and rigorous response time. Therefore,
top-performing deep models of high depth and complexity are not well suited for
these applications with the limitations on the inference time. In order to
further improve the neural networks' performance given the time and
computational limitations, we propose an approach that exploits a cumbersome
net to help train the lightweight net for prediction. We dub the whole process
rocket launching, where the cumbersome booster net is used to guide the
learning of the target light net throughout the whole training process. We
analyze different loss functions aiming at pushing the light net to behave
similarly to the booster net, and adopt the loss with best performance in our
experiments. We use one technique called gradient block to improve the
performance of the light net and booster net further. Experiments on benchmark
datasets and real-life industrial advertisement data present that our light
model can get performance only previously achievable with more complex models.
| 1 | 0 | 0 | 1 | 0 | 0 |
The CMS HGCAL detector for HL-LHC upgrade | The High Luminosity LHC (HL-LHC) will integrate 10 times more luminosity than
the LHC, posing significant challenges for radiation tolerance and event pileup
on detectors, especially for forward calorimetry, and hallmarks the issue for
future colliders. As part of its HL-LHC upgrade program, the CMS collaboration
is designing a High Granularity Calorimeter to replace the existing endcap
calorimeters. It features unprecedented transverse and longitudinal
segmentation for both electromagnetic (ECAL) and hadronic (HCAL) compartments.
This will facilitate particle-flow calorimetry, where the fine structure of
showers can be measured and used to enhance pileup rejection and particle
identification, whilst still achieving good energy resolution. The ECAL and a
large fraction of HCAL will be based on hexagonal silicon sensors of
0.5-1cm$^{2}$ cell size, with the remainder of the HCAL based on
highly-segmented scintillators with SiPM readout. The intrinsic high-precision
timing capabilities of the silicon sensors will add an extra dimension to event
reconstruction, especially in terms of pileup rejection. An overview of the
HGCAL project is presented, covering motivation, engineering design, readout
and trigger concepts, and performance (simulated and from beam tests).
| 0 | 1 | 0 | 0 | 0 | 0 |
Complete Classification of Generalized Santha-Vazirani Sources | Let $\mathcal{F}$ be a finite alphabet and $\mathcal{D}$ be a finite set of
distributions over $\mathcal{F}$. A Generalized Santha-Vazirani (GSV) source of
type $(\mathcal{F}, \mathcal{D})$, introduced by Beigi, Etesami and Gohari
(ICALP 2015, SICOMP 2017), is a random sequence $(F_1, \dots, F_n)$ in
$\mathcal{F}^n$, where $F_i$ is a sample from some distribution $d \in
\mathcal{D}$ whose choice may depend on $F_1, \dots, F_{i-1}$.
We show that all GSV source types $(\mathcal{F}, \mathcal{D})$ fall into one
of three categories: (1) non-extractable; (2) extractable with error
$n^{-\Theta(1)}$; (3) extractable with error $2^{-\Omega(n)}$. This rules out
other error rates like $1/\log n$ or $2^{-\sqrt{n}}$.
We provide essentially randomness-optimal extraction algorithms for
extractable sources. Our algorithm for category (2) sources extracts with error
$\varepsilon$ from $n = \mathrm{poly}(1/\varepsilon)$ samples in time linear in
$n$. Our algorithm for category (3) sources extracts $m$ bits with error
$\varepsilon$ from $n = O(m + \log 1/\varepsilon)$ samples in time
$\min\{O(nm2^m),n^{O(\lvert\mathcal{F}\rvert)}\}$.
We also give algorithms for classifying a GSV source type $(\mathcal{F},
\mathcal{D})$: Membership in category (1) can be decided in $\mathrm{NP}$,
while membership in category (3) is polynomial-time decidable.
| 1 | 0 | 0 | 0 | 0 | 0 |
Hermann Hankel's "On the general theory of motion of fluids", an essay including an English translation of the complete Preisschrift from 1861 | The present is a companion paper to "A contemporary look at Hermann Hankel's
1861 pioneering work on Lagrangian fluid dynamics" by Frisch, Grimberg and
Villone (2017). Here we present the English translation of the 1861 prize
manuscript from Göttingen University "Zur allgemeinen Theorie der Bewegung
der Flüssigkeiten" (On the general theory of the motion of the fluids) of
Hermann Hankel (1839-1873), which was originally submitted in Latin and then
translated into German by the Author for publication. We also provide the
English translation of two important reports on the manuscript, one written by
Bernhard Riemann and the other by Wilhelm Eduard Weber, during the assessment
process for the prize. Finally we give a short biography of Hermann Hankel with
his complete bibliography.
| 0 | 1 | 1 | 0 | 0 | 0 |
Targeted Damage to Interdependent Networks | The giant mutually connected component (GMCC) of an interdependent or
multiplex network collapses with a discontinuous hybrid transition under random
damage to the network. If the nodes to be damaged are selected in a targeted
way, the collapse of the GMCC may occur significantly sooner. Finding the
minimal damage set which destroys the largest mutually connected component of a
given interdependent network is a computationally prohibitive simultaneous
optimization problem. We introduce a simple heuristic strategy -- Effective
Multiplex Degree -- for targeted attack on interdependent networks that
leverages the indirect damage inherent in multiplex networks to achieve a
damage set smaller than that found by any other non computationally intensive
algorithm. We show that the intuition from single layer networks that decycling
(damage of the $2$-core) is the most effective way to destroy the giant
component, does not carry over to interdependent networks, and in fact such
approaches are worse than simply removing the highest degree nodes.
| 1 | 0 | 0 | 0 | 0 | 0 |
The limit of the Hermitian-Yang-Mills flow on reflexive sheaves | In this paper, we study the asymptotic behavior of the Hermitian-Yang-Mills
flow on a reflexive sheaf. We prove that the limiting reflexive sheaf is
isomorphic to the double dual of the graded sheaf associated to the
Harder-Narasimhan-Seshadri filtration, this answers a question by Bando and
Siu.
| 0 | 0 | 1 | 0 | 0 | 0 |
High-accuracy phase-field models for brittle fracture based on a new family of degradation functions | Phase-field approaches to fracture based on energy minimization principles
have been rapidly gaining popularity in recent years, and are particularly
well-suited for simulating crack initiation and growth in complex fracture
networks. In the phase-field framework, the surface energy associated with
crack formation is calculated by evaluating a functional defined in terms of a
scalar order parameter and its gradients, which in turn describe the fractures
in a diffuse sense following a prescribed regularization length scale. Imposing
stationarity of the total energy leads to a coupled system of partial
differential equations, one enforcing stress equilibrium and another governing
phase-field evolution. The two equations are coupled through an energy
degradation function that models the loss of stiffness in the bulk material as
it undergoes damage. In the present work, we introduce a new parametric family
of degradation functions aimed at increasing the accuracy of phase-field models
in predicting critical loads associated with crack nucleation as well as the
propagation of existing fractures. An additional goal is the preservation of
linear elastic response in the bulk material prior to fracture. Through the
analysis of several numerical examples, we demonstrate the superiority of the
proposed family of functions to the classical quadratic degradation function
that is used most often in the literature.
| 0 | 1 | 1 | 0 | 0 | 0 |
Straggler Mitigation in Distributed Optimization Through Data Encoding | Slow running or straggler tasks can significantly reduce computation speed in
distributed computation. Recently, coding-theory-inspired approaches have been
applied to mitigate the effect of straggling, through embedding redundancy in
certain linear computational steps of the optimization algorithm, thus
completing the computation without waiting for the stragglers. In this paper,
we propose an alternate approach where we embed the redundancy directly in the
data itself, and allow the computation to proceed completely oblivious to
encoding. We propose several encoding schemes, and demonstrate that popular
batch algorithms, such as gradient descent and L-BFGS, applied in a
coding-oblivious manner, deterministically achieve sample path linear
convergence to an approximate solution of the original problem, using an
arbitrarily varying subset of the nodes at each iteration. Moreover, this
approximation can be controlled by the amount of redundancy and the number of
nodes used in each iteration. We provide experimental results demonstrating the
advantage of the approach over uncoded and data replication strategies.
| 1 | 0 | 0 | 1 | 0 | 0 |
Inference Trees: Adaptive Inference with Exploration | We introduce inference trees (ITs), a new class of inference methods that
build on ideas from Monte Carlo tree search to perform adaptive sampling in a
manner that balances exploration with exploitation, ensures consistency, and
alleviates pathologies in existing adaptive methods. ITs adaptively sample from
hierarchical partitions of the parameter space, while simultaneously learning
these partitions in an online manner. This enables ITs to not only identify
regions of high posterior mass, but also maintain uncertainty estimates to
track regions where significant posterior mass may have been missed. ITs can be
based on any inference method that provides a consistent estimate of the
marginal likelihood. They are particularly effective when combined with
sequential Monte Carlo, where they capture long-range dependencies and yield
improvements beyond proposal adaptation alone.
| 0 | 0 | 0 | 1 | 0 | 0 |
Application of backpropagation neural networks to both stages of fingerprinting based WIPS | We propose a scheme to employ backpropagation neural networks (BPNNs) for
both stages of fingerprinting-based indoor positioning using WLAN/WiFi signal
strengths (FWIPS): radio map construction during the offline stage, and
localization during the online stage. Given a training radio map (TRM), i.e., a
set of coordinate vectors and associated WLAN/WiFi signal strengths of the
available access points, a BPNN can be trained to output the expected signal
strengths for any input position within the region of interest (BPNN-RM). This
can be used to provide a continuous representation of the radio map and to
filter, densify or decimate a discrete radio map. Correspondingly, the TRM can
also be used to train another BPNN to output the expected position within the
region of interest for any input vector of recorded signal strengths and thus
carry out localization (BPNN-LA).Key aspects of the design of such artificial
neural networks for a specific application are the selection of design
parameters like the number of hidden layers and nodes within the network, and
the training procedure. Summarizing extensive numerical simulations, based on
real measurements in a testbed, we analyze the impact of these design choices
on the performance of the BPNN and compare the results in particular to those
obtained using the $k$ nearest neighbors ($k$NN) and weighted $k$ nearest
neighbors approaches to FWIPS.
| 1 | 0 | 0 | 1 | 0 | 0 |
Bayesian Bootstraps for Massive Data | Recently, two scalable adaptations of the bootstrap have been proposed: the
bag of little bootstraps (BLB; Kleiner et al., 2014) and the subsampled double
bootstrap (SDB; Sengupta et al., 2016). In this paper, we introduce Bayesian
bootstrap analogues to the BLB and SDB that have similar theoretical and
computational properties, a strategy to perform lossless inference for a class
of functionals of the Bayesian bootstrap, and briefly discuss extensions for
Dirichlet Processes.
| 0 | 0 | 0 | 1 | 0 | 0 |
Show, Adapt and Tell: Adversarial Training of Cross-domain Image Captioner | Impressive image captioning results are achieved in domains with plenty of
training image and sentence pairs (e.g., MSCOCO). However, transferring to a
target domain with significant domain shifts but no paired training data
(referred to as cross-domain image captioning) remains largely unexplored. We
propose a novel adversarial training procedure to leverage unpaired data in the
target domain. Two critic networks are introduced to guide the captioner,
namely domain critic and multi-modal critic. The domain critic assesses whether
the generated sentences are indistinguishable from sentences in the target
domain. The multi-modal critic assesses whether an image and its generated
sentence are a valid pair. During training, the critics and captioner act as
adversaries -- captioner aims to generate indistinguishable sentences, whereas
critics aim at distinguishing them. The assessment improves the captioner
through policy gradient updates. During inference, we further propose a novel
critic-based planning method to select high-quality sentences without
additional supervision (e.g., tags). To evaluate, we use MSCOCO as the source
domain and four other datasets (CUB-200-2011, Oxford-102, TGIF, and Flickr30k)
as the target domains. Our method consistently performs well on all datasets.
In particular, on CUB-200-2011, we achieve 21.8% CIDEr-D improvement after
adaptation. Utilizing critics during inference further gives another 4.5%
boost.
| 1 | 0 | 0 | 0 | 0 | 0 |
Faster Fuzzing: Reinitialization with Deep Neural Models | We improve the performance of the American Fuzzy Lop (AFL) fuzz testing
framework by using Generative Adversarial Network (GAN) models to reinitialize
the system with novel seed files. We assess performance based on the temporal
rate at which we produce novel and unseen code paths. We compare this approach
to seed file generation from a random draw of bytes observed in the training
seed files. The code path lengths and variations were not sufficiently diverse
to fully replace AFL input generation. However, augmenting native AFL with
these additional code paths demonstrated improvements over AFL alone.
Specifically, experiments showed the GAN was faster and more effective than the
LSTM and out-performed a random augmentation strategy, as measured by the
number of unique code paths discovered. GAN helps AFL discover 14.23% more code
paths than the random strategy in the same amount of CPU time, finds 6.16% more
unique code paths, and finds paths that are on average 13.84% longer. Using GAN
shows promise as a reinitialization strategy for AFL to help the fuzzer
exercise deep paths in software.
| 1 | 0 | 0 | 0 | 0 | 0 |
Contego: An Adaptive Framework for Integrating Security Tasks in Real-Time Systems | Embedded real-time systems (RTS) are pervasive. Many modern RTS are exposed
to unknown security flaws, and threats to RTS are growing in both number and
sophistication. However, until recently, cyber-security considerations were an
afterthought in the design of such systems. Any security mechanisms integrated
into RTS must (a) co-exist with the real- time tasks in the system and (b)
operate without impacting the timing and safety constraints of the control
logic. We introduce Contego, an approach to integrating security tasks into RTS
without affecting temporal requirements. Contego is specifically designed for
legacy systems, viz., the real-time control systems in which major alterations
of the system parameters for constituent tasks is not always feasible. Contego
combines the concept of opportunistic execution with hierarchical scheduling to
maintain compatibility with legacy systems while still providing flexibility by
allowing security tasks to operate in different modes. We also define a metric
to measure the effectiveness of such integration. We evaluate Contego using
synthetic workloads as well as with an implementation on a realistic embedded
platform (an open- source ARM CPU running real-time Linux).
| 1 | 0 | 0 | 0 | 0 | 0 |
Second Order Analysis for Joint Source-Channel Coding with Markovian Source | We derive the second order rates of joint source-channel coding, whose source
obeys an irreducible and ergodic Markov process when the channel is a discrete
memoryless, while a previous study solved it only in a special case. We also
compare the joint source-channel scheme with the separation scheme in the
second order regime while a previous study made a notable comparison only with
numerical calculation. To make these two notable progress, we introduce two
kinds of new distribution families, switched Gaussian convolution distribution
and *-product distribution, which are defined by modifying the Gaussian
distribution.
| 1 | 0 | 1 | 0 | 0 | 0 |
Interface currents and magnetization in singlet-triplet superconducting heterostructures: Role of chiral and helical domains | Chiral and helical domain walls are generic defects of topological
spin-triplet superconductors. We study theoretically the magnetic and transport
properties of superconducting singlet-triplet-singlet heterostructure as a
function of the phase difference between the singlet leads in the presence of
chiral and helical domains inside the spin-triplet region. The local inversion
symmetry breaking at the singlet-triplet interface allows the emergence of a
static phase-controlled magnetization, and generally yields both spin and
charge currents flowing along the edges. The parity of the domain wall number
affects the relative orientation of the interface moments and currents, while
in some cases the domain walls themselves contribute to spin and charge
transport. We demonstrate that singlet-triplet heterostructures are a generic
prototype to generate and control non-dissipative spin and charge effects,
putting them in a broader class of systems exhibiting spin-Hall, anomalous Hall
effects and similar phenomena. Features of the electron transport and magnetic
effects at the interfaces can be employed to assess the presence of domains in
chiral/helical superconductors.
| 0 | 1 | 0 | 0 | 0 | 0 |
Implicit Weight Uncertainty in Neural Networks | Modern neural networks tend to be overconfident on unseen, noisy or
incorrectly labelled data and do not produce meaningful uncertainty measures.
Bayesian deep learning aims to address this shortcoming with variational
approximations (such as Bayes by Backprop or Multiplicative Normalising Flows).
However, current approaches have limitations regarding flexibility and
scalability. We introduce Bayes by Hypernet (BbH), a new method of variational
approximation that interprets hypernetworks as implicit distributions. It
naturally uses neural networks to model arbitrarily complex distributions and
scales to modern deep learning architectures. In our experiments, we
demonstrate that our method achieves competitive accuracies and predictive
uncertainties on MNIST and a CIFAR5 task, while being the most robust against
adversarial attacks.
| 1 | 0 | 0 | 1 | 0 | 0 |
A systematic analysis of the XMM-Newton background: III. Impact of the magnetospheric environment | A detailed characterization of the particle induced background is fundamental
for many of the scientific objectives of the Athena X-ray telescope, thus an
adequate knowledge of the background that will be encountered by Athena is
desirable. Current X-ray telescopes have shown that the intensity of the
particle induced background can be highly variable. Different regions of the
magnetosphere can have very different environmental conditions, which can, in
principle, differently affect the particle induced background detected by the
instruments. We present results concerning the influence of the magnetospheric
environment on the background detected by EPIC instrument onboard XMM-Newton
through the estimate of the variation of the in-Field-of-View background excess
along the XMM-Newton orbit. An important contribution to the XMM background,
which may affect the Athena background as well, comes from soft proton flares.
Along with the flaring component a low-intensity component is also present. We
find that both show modest variations in the different magnetozones and that
the soft proton component shows a strong trend with the distance from Earth.
| 0 | 1 | 0 | 0 | 0 | 0 |
DeepPermNet: Visual Permutation Learning | We present a principled approach to uncover the structure of visual data by
solving a novel deep learning task coined visual permutation learning. The goal
of this task is to find the permutation that recovers the structure of data
from shuffled versions of it. In the case of natural images, this task boils
down to recovering the original image from patches shuffled by an unknown
permutation matrix. Unfortunately, permutation matrices are discrete, thereby
posing difficulties for gradient-based methods. To this end, we resort to a
continuous approximation of these matrices using doubly-stochastic matrices
which we generate from standard CNN predictions using Sinkhorn iterations.
Unrolling these iterations in a Sinkhorn network layer, we propose DeepPermNet,
an end-to-end CNN model for this task. The utility of DeepPermNet is
demonstrated on two challenging computer vision problems, namely, (i) relative
attributes learning and (ii) self-supervised representation learning. Our
results show state-of-the-art performance on the Public Figures and OSR
benchmarks for (i) and on the classification and segmentation tasks on the
PASCAL VOC dataset for (ii).
| 1 | 0 | 0 | 0 | 0 | 0 |
ADE String Chains and Mirror Symmetry | 6d superconformal field theories (SCFTs) are the SCFTs in the highest
possible dimension. They can be geometrically engineered in F-theory by
compactifying on non-compact elliptic Calabi-Yau manifolds. In this paper we
focus on the class of SCFTs whose base geometry is determined by $-2$ curves
intersecting according to ADE Dynkin diagrams and derive the corresponding
mirror Calabi-Yau manifold. The mirror geometry is uniquely determined in terms
of the mirror curve which has also an interpretation in terms of the
Seiberg-Witten curve of the four-dimensional theory arising from torus
compactification. Adding the affine node of the ADE quiver to the base
geometry, we connect to recent results on SYZ mirror symmetry for the $A$ case
and provide a physical interpretation in terms of little string theory. Our
results, however, go beyond this case as our construction naturally covers the
$D$ and $E$ cases as well.
| 0 | 0 | 1 | 0 | 0 | 0 |
(non)-automaticity of completely multiplicative sequences having negligible many non-trivial prime factors | In this article we consider the completely multiplicative sequences $(a_n)_{n
\in \mathbf{N}}$ defined on a field $\mathbf{K}$ and satisfying $$\sum_{p| p
\leq n, a_p \neq 1, p \in \mathbf{P}}\frac{1}{p}<\infty,$$ where $\mathbf{P}$
is the set of prime numbers. We prove that if such sequences are automatic then
they cannot have infinitely many prime numbers $p$ such that $a_{p}\neq 1$.
Using this fact, we prove that if a completely multiplicative sequence
$(a_n)_{n \in \mathbf{N}}$, vanishing or not, can be written in the form
$a_n=b_n\chi_n$ such that $(b_n)_{n \in \mathbf{N}}$ is a non ultimately
periodic, completely multiplicative automatic sequence satisfying the above
condition, and $(\chi_n)_{n \in \mathbf{N}}$ is a Dirichlet character or a
constant sequence, then there exists only one prime number $p$ such that $b_p
\neq 1$ or $0$.
| 0 | 0 | 1 | 0 | 0 | 0 |
Timing Aware Dummy Metal Fill Methodology | In this paper, we analyzed parasitic coupling capacitance coming from dummy
metal fill and its impact on timing. Based on the modeling, we proposed two
approaches to minimize the timing impact from dummy metal fill. The first
approach applies more spacing between critical nets and metal fill, while the
second approach leverages the shielding effects of reference nets. Experimental
results show consistent improvement compared to traditional metal fill method.
| 1 | 0 | 0 | 0 | 0 | 0 |
Asymptotic efficiency of the proportional compensation scheme for a large number of producers | We consider a manager, who allocates some fixed total payment amount between
$N$ rational agents in order to maximize the aggregate production. The profit
of $i$-th agent is the difference between the compensation (reward) obtained
from the manager and the production cost. We compare (i) the \emph{normative}
compensation scheme, where the manager enforces the agents to follow an optimal
cooperative strategy; (ii) the \emph{linear piece rates} compensation scheme,
where the manager announces an optimal reward per unit good; (iii) the
\emph{proportional} compensation scheme, where agent's reward is proportional
to his contribution to the total output. Denoting the correspondent total
production levels by $s^*$, $\hat s$ and $\overline s$ respectively, where the
last one is related to the unique Nash equilibrium, we examine the limits of
the prices of anarchy $\mathscr A_N=s^*/\overline s$, $\mathscr A_N'=\hat
s/\overline s$ as $N\to\infty$. These limits are calculated for the cases of
identical convex costs with power asymptotics at the origin, and for power
costs, corresponding to the Coob-Douglas and generalized CES production
functions with decreasing returns to scale. Our results show that
asymptotically no performance is lost in terms of $\mathscr A'_N$, and in terms
of $\mathscr A_N$ the loss does not exceed $31\%$.
| 1 | 0 | 0 | 0 | 0 | 0 |
Non-equilibrium statistical mechanics of continuous attractors | Continuous attractors have been used to understand recent neuroscience
experiments where persistent activity patterns encode internal representations
of external attributes like head direction or spatial location. However, the
conditions under which the emergent bump of neural activity in such networks
can be manipulated by space and time-dependent external sensory or motor
signals are not understood. Here, we find fundamental limits on how rapidly
internal representations encoded along continuous attractors can be updated by
an external signal. We apply these results to place cell networks to derive a
velocity-dependent non-equilibrium memory capacity in neural networks.
| 0 | 0 | 0 | 0 | 1 | 0 |
Some results on the existence of t-all-or-nothing transforms over arbitrary alphabets | A $(t, s, v)$-all-or-nothing transform is a bijective mapping defined on
$s$-tuples over an alphabet of size $v$, which satisfies the condition that the
values of any $t$ input co-ordinates are completely undetermined, given only
the values of any $s-t$ output co-ordinates. The main question we address in
this paper is: for which choices of parameters does a $(t, s,
v)$-all-or-nothing transform (AONT) exist? More specifically, if we fix $t$ and
$v$, we want to determine the maximum integer $s$ such that a $(t, s, v)$-AONT
exists. We mainly concentrate on the case $t=2$ for arbitrary values of $v$,
where we obtain various necessary as well as sufficient conditions for
existence of these objects. We consider both linear and general (linear or
nonlinear) AONT. We also show some connections between AONT, orthogonal arrays
and resilient functions.
| 1 | 0 | 1 | 0 | 0 | 0 |
Exhaustive Exploration of the Failure-oblivious Computing Search Space | High-availability of software systems requires automated handling of crashes
in presence of errors. Failure-oblivious computing is one technique that aims
to achieve high availability. We note that failure-obliviousness has not been
studied in depth yet, and there is very few study that helps understand why
failure-oblivious techniques work. In order to make failure-oblivious computing
to have an impact in practice, we need to deeply understand failure-oblivious
behaviors in software. In this paper, we study, design and perform an
experiment that analyzes the size and the diversity of the failure-oblivious
behaviors. Our experiment consists of exhaustively computing the search space
of 16 field failures of large-scale open-source Java software. The outcome of
this experiment is a much better understanding of what really happens when
failure-oblivious computing is used, and this opens new promising research
directions.
| 1 | 0 | 0 | 0 | 0 | 0 |
Theoretical Accuracy in Cosmological Growth Estimation | We elucidate the importance of the consistent treatment of gravity-model
specific non-linearities when estimating the growth of cosmological structures
from redshift space distortions (RSD). Within the context of standard
perturbation theory (SPT), we compare the predictions of two theoretical
templates with redshift space data from COLA (COmoving Lagrangian Acceleration)
simulations in the normal branch of DGP gravity (nDGP) and General Relativity
(GR). Using COLA for these comparisons is validated using a suite of full
N-body simulations for the same theories. The two theoretical templates
correspond to the standard general relativistic perturbation equations and
those same equations modelled within nDGP. Gravitational clustering non-linear
effects are accounted for by modelling the power spectrum up to one loop order
and redshift space clustering anisotropy is modelled using the Taruya,
Nishimichi and Saito (TNS) RSD model. Using this approach, we attempt to
recover the simulation's fiducial logarithmic growth parameter $f$. By
assigning the simulation data with errors representing an idealised survey with
a volume of $10\mbox{Gpc}^3/h^3$, we find the GR template is unable to recover
fiducial $f$ to within 1$\sigma$ at $z=1$ when we match the data up to $k_{\rm
max}=0.195h$/Mpc. On the other hand, the DGP template recovers the fiducial
value within $1\sigma$. Further, we conduct the same analysis for sets of mock
data generated for generalised models of modified gravity using SPT, where
again we analyse the GR template's ability to recover the fiducial value. We
find that for models with enhanced gravitational non-linearity, the theoretical
bias of the GR template becomes significant for stage IV surveys. Thus, we show
that for the future large data volume galaxy surveys, the self-consistent
modelling of non-GR gravity scenarios will be crucial in constraining theory
parameters.
| 0 | 1 | 0 | 0 | 0 | 0 |
Model-Robust Counterfactual Prediction Method | We develop a novel method for counterfactual analysis based on observational
data using prediction intervals for units under different exposures. Unlike
methods that target heterogeneous or conditional average treatment effects of
an exposure, the proposed approach aims to take into account the irreducible
dispersions of counterfactual outcomes so as to quantify the relative impact of
different exposures. The prediction intervals are constructed in a
distribution-free and model-robust manner based on the conformal prediction
approach. The computational obstacles to this approach are circumvented by
leveraging properties of a tuning-free method that learns sparse additive
predictor models for counterfactual outcomes. The method is illustrated using
both real and synthetic data.
| 0 | 0 | 1 | 1 | 0 | 0 |
Exponentiated Generalized Pareto Distribution: Properties and applications towards Extreme Value Theory | The Generalized Pareto Distribution (GPD) plays a central role in modelling
heavy tail phenomena in many applications. Applying the GPD to actual datasets
however is a non-trivial task. One common way suggested in the literature to
investigate the tail behaviour is to take logarithm to the original dataset in
order to reduce the sample variability. Inspired by this, we propose and study
the Exponentiated Generalized Pareto Distribution (exGPD), which is created via
log-transform of the GPD variable. After introducing the exGPD we derive
various distributional quantities, including the moment generating function,
tail risk measures. As an application we also develop a plot as an alternative
to the Hill plot to identify the tail index of heavy tailed datasets, based on
the moment matching for the exGPD. Various numerical analyses with both
simulated and actual datasets show that the proposed plot works well.
| 0 | 0 | 1 | 1 | 0 | 0 |
Learning with Average Top-k Loss | In this work, we introduce the {\em average top-$k$} (\atk) loss as a new
aggregate loss for supervised learning, which is the average over the $k$
largest individual losses over a training dataset. We show that the \atk loss
is a natural generalization of the two widely used aggregate losses, namely the
average loss and the maximum loss, but can combine their advantages and
mitigate their drawbacks to better adapt to different data distributions.
Furthermore, it remains a convex function over all individual losses, which can
lead to convex optimization problems that can be solved effectively with
conventional gradient-based methods. We provide an intuitive interpretation of
the \atk loss based on its equivalent effect on the continuous individual loss
functions, suggesting that it can reduce the penalty on correctly classified
data. We further give a learning theory analysis of \matk learning on the
classification calibration of the \atk loss and the error bounds of \atk-SVM.
We demonstrate the applicability of minimum average top-$k$ learning for binary
classification and regression using synthetic and real datasets.
| 1 | 0 | 0 | 1 | 0 | 0 |
Reflexive polytopes arising from perfect graphs | Reflexive polytopes form one of the distinguished classes of lattice
polytopes. Especially reflexive polytopes which possess the integer
decomposition property are of interest. In the present paper, by virtue of the
algebraic technique on Grönbner bases, a new class of reflexive polytopes
which possess the integer decomposition property and which arise from perfect
graphs will be presented. Furthermore, the Ehrhart $\delta$-polynomials of
these polytopes will be studied.
| 0 | 0 | 1 | 0 | 0 | 0 |
Meta Networks | Neural networks have been successfully applied in applications with a large
amount of labeled data. However, the task of rapid generalization on new
concepts with small training data while preserving performances on previously
learned ones still presents a significant challenge to neural network models.
In this work, we introduce a novel meta learning method, Meta Networks
(MetaNet), that learns a meta-level knowledge across tasks and shifts its
inductive biases via fast parameterization for rapid generalization. When
evaluated on Omniglot and Mini-ImageNet benchmarks, our MetaNet models achieve
a near human-level performance and outperform the baseline approaches by up to
6% accuracy. We demonstrate several appealing properties of MetaNet relating to
generalization and continual learning.
| 1 | 0 | 0 | 1 | 0 | 0 |
Variable selection for clustering with Gaussian mixture models: state of the art | The mixture models have become widely used in clustering, given its
probabilistic framework in which its based, however, for modern databases that
are characterized by their large size, these models behave disappointingly in
setting out the model, making essential the selection of relevant variables for
this type of clustering. After recalling the basics of clustering based on a
model, this article will examine the variable selection methods for model-based
clustering, as well as presenting opportunities for improvement of these
methods.
| 1 | 0 | 0 | 1 | 0 | 0 |
Analysing Magnetism Using Scanning SQUID Microscopy | Scanning superconducting quantum interference device microscopy (SSM) is a
scanning probe technique that images local magnetic flux, which allows for
mapping of magnetic fields with high field and spatial accuracy. Many studies
involving SSM have been published in the last decades, using SSM to make
qualitative statements about magnetism. However, quantitative analysis using
SSM has received less attention. In this work, we discuss several aspects of
interpreting SSM images and methods to improve quantitative analysis. First, we
analyse the spatial resolution and how it depends on several factors. Second,
we discuss the analysis of SSM scans and the information obtained from the SSM
data. Using simulations, we show how signals evolve as a function of changing
scan height, SQUID loop size, magnetization strength and orientation. We also
investigated 2-dimensional autocorrelation analysis to extract information
about the size, shape and symmetry of magnetic features. Finally, we provide an
outlook on possible future applications and improvements.
| 0 | 1 | 0 | 0 | 0 | 0 |
Algorithms for solving optimization problems arising from deep neural net models: nonsmooth problems | Machine Learning models incorporating multiple layered learning networks have
been seen to provide effective models for various classification problems. The
resulting optimization problem to solve for the optimal vector minimizing the
empirical risk is, however, highly nonconvex. This alone presents a challenge
to application and development of appropriate optimization algorithms for
solving the problem. However, in addition, there are a number of interesting
problems for which the objective function is non- smooth and nonseparable. In
this paper, we summarize the primary challenges involved, the state of the art,
and present some numerical results on an interesting and representative class
of problems.
| 0 | 0 | 0 | 1 | 0 | 0 |
On the essential self-adjointness of singular sub-Laplacians | We prove a general essential self-adjointness criterion for sub-Laplacians on
complete sub-Riemannian manifolds, defined with respect to singular measures.
As a consequence, we show that the intrinsic sub-Laplacian (i.e. defined w.r.t.
Popp's measure) is essentially self-adjoint on the equiregular connected
components of a sub-Riemannian manifold. This result holds under mild
regularity assumptions on the singular region, and when the latter does not
contain characteristic points.
| 0 | 0 | 1 | 0 | 0 | 0 |
Are Saddles Good Enough for Deep Learning? | Recent years have seen a growing interest in understanding deep neural
networks from an optimization perspective. It is understood now that converging
to low-cost local minima is sufficient for such models to become effective in
practice. However, in this work, we propose a new hypothesis based on recent
theoretical findings and empirical studies that deep neural network models
actually converge to saddle points with high degeneracy. Our findings from this
work are new, and can have a significant impact on the development of gradient
descent based methods for training deep networks. We validated our hypotheses
using an extensive experimental evaluation on standard datasets such as MNIST
and CIFAR-10, and also showed that recent efforts that attempt to escape
saddles finally converge to saddles with high degeneracy, which we define as
`good saddles'. We also verified the famous Wigner's Semicircle Law in our
experimental results.
| 1 | 0 | 0 | 1 | 0 | 0 |
Monotonicity and enclosure methods for the p-Laplace equation | We show that the convex hull of a monotone perturbation of a homogeneous
background conductivity in the $p$-conductivity equation is determined by
knowledge of the nonlinear Dirichlet-Neumann operator. We give two independent
proofs, one of which is based on the monotonicity method and the other on the
enclosure method. Our results are constructive and require no jump or
smoothness properties on the conductivity perturbation or its support.
| 0 | 0 | 1 | 0 | 0 | 0 |
Tension and chemical efficiency of Myosin-II motors | Recent experiments demonstrate that molecular motors from the Myosin II
family serve as cross-links inducing active tension in the cytoskeletal
network. Here we revise the Brownian ratchet model, previously studied in the
context of active transport along polymer tracks, in setups resembling a motor
in a polymer network, also taking into account the effect of electrostatic
changes in the motor heads. We explore important mechanical quantities and show
that such a model is also capable of mechanosensing. Finally, we introduce a
novel efficiency based on excess heat production by the chemical cycle which is
directly related to the active tension the motor exerts. The chemical
efficiencies differ considerably for motors with a different number of heads,
while their mechanical properties remain qualitatively similar. For motors with
a small number of heads, the chemical efficiency is maximal when they are
frustrated, a trait that is not found in larger motors.
| 0 | 1 | 0 | 0 | 0 | 0 |
Token-based Function Computation with Memory | In distributed function computation, each node has an initial value and the
goal is to compute a function of these values in a distributed manner. In this
paper, we propose a novel token-based approach to compute a wide class of
target functions to which we refer as "Token-based function Computation with
Memory" (TCM) algorithm. In this approach, node values are attached to tokens
and travel across the network. Each pair of travelling tokens would coalesce
when they meet, forming a token with a new value as a function of the original
token values. In contrast to the Coalescing Random Walk (CRW) algorithm, where
token movement is governed by random walk, meeting of tokens in our scheme is
accelerated by adopting a novel chasing mechanism. We proved that, compared to
the CRW algorithm, the TCM algorithm results in a reduction of time complexity
by a factor of at least $\sqrt{n/\log(n)}$ in Erdös-Renyi and complete
graphs, and by a factor of $\log(n)/\log(\log(n))$ in torus networks.
Simulation results show that there is at least a constant factor improvement in
the message complexity of TCM algorithm in all considered topologies.
Robustness of the CRW and TCM algorithms in the presence of node failure is
analyzed. We show that their robustness can be improved by running multiple
instances of the algorithms in parallel.
| 1 | 0 | 0 | 1 | 0 | 0 |
Simple property of heterogeneous aspiration dynamics: Beyond weak selection | How individuals adapt their behavior in cultural evolution remains elusive.
Theoretical studies have shown that the update rules chosen to model individual
decision making can dramatically modify the evolutionary outcome of the
population as a whole. This hints at the complexities of considering the
personality of individuals in a population, where each one uses its own rule.
Here, we investigate whether and how heterogeneity in the rules of behavior
update alters the evolutionary outcome. We assume that individuals update
behaviors by aspiration-based self-evaluation and they do so in their own ways.
Under weak selection, we analytically reveal a simple property that holds for
any two-strategy multi-player games in well-mixed populations and on regular
graphs: the evolutionary outcome in a population with heterogeneous update
rules is the weighted average of the outcomes in the corresponding homogeneous
populations, and the associated weights are the frequencies of each update rule
in the heterogeneous population. Beyond weak selection, we show that this
property holds for public goods games. Our finding implies that heterogeneous
aspiration dynamics is additive. This additivity greatly reduces the complexity
induced by the underlying individual heterogeneity. Our work thus provides an
efficient method to calculate evolutionary outcomes under heterogeneous update
rules.
| 0 | 0 | 0 | 0 | 1 | 0 |
Warm dark matter and the ionization history of the Universe | In warm dark matter scenarios structure formation is suppressed on small
scales with respect to the cold dark matter case, reducing the number of
low-mass halos and the fraction of ionized gas at high redshifts and thus,
delaying reionization. This has an impact on the ionization history of the
Universe and measurements of the optical depth to reionization, of the
evolution of the global fraction of ionized gas and of the thermal history of
the intergalactic medium, can be used to set constraints on the mass of the
dark matter particle. However, the suppression of the fraction of ionized
medium in these scenarios can be partly compensated by varying other
parameters, as the ionization efficiency or the minimum mass for which halos
can host star-forming galaxies. Here we use different data sets regarding the
ionization and thermal histories of the Universe and, taking into account the
degeneracies from several astrophysical parameters, we obtain a lower bound on
the mass of thermal warm dark matter candidates of $m_X > 1.3$ keV, or $m_s >
5.5$ keV for the case of sterile neutrinos non-resonantly produced in the early
Universe, both at 90\% confidence level.
| 0 | 1 | 0 | 0 | 0 | 0 |
High quality factor manganese-doped aluminum lumped-element kinetic inductance detectors sensitive to frequencies below 100 GHz | Aluminum lumped-element kinetic inductance detectors (LEKIDs) sensitive to
millimeter-wave photons have been shown to exhibit high quality factors, making
them highly sensitive and multiplexable. The superconducting gap of aluminum
limits aluminum LEKIDs to photon frequencies above 100 GHz. Manganese-doped
aluminum (Al-Mn) has a tunable critical temperature and could therefore be an
attractive material for LEKIDs sensitive to frequencies below 100 GHz if the
internal quality factor remains sufficiently high when manganese is added to
the film. To investigate, we measured some of the key properties of Al-Mn
LEKIDs. A prototype eight-element LEKID array was fabricated using a 40 nm
thick film of Al-Mn deposited on a 500 {\mu}m thick high-resistivity,
float-zone silicon substrate. The manganese content was 900 ppm, the measured
$T_c = 694\pm1$ mK, and the resonance frequencies were near 150 MHz. Using
measurements of the forward scattering parameter $S_{21}$ at various bath
temperatures between 65 and 250 mK, we determined that the Al-Mn LEKIDs we
fabricated have internal quality factors greater than $2 \times 10^5$, which is
high enough for millimeter-wave astrophysical observations. In the dark
conditions under which these devices were measured, the fractional frequency
noise spectrum shows a shallow slope that depends on bath temperature and probe
tone amplitude, which could be two-level system noise. The anticipated white
photon noise should dominate this level of low-frequency noise when the
detectors are illuminated with millimeter-waves in future measurements. The
LEKIDs responded to light pulses from a 1550 nm light-emitting diode, and we
used these light pulses to determine that the quasiparticle lifetime is 60
{\mu}s.
| 0 | 1 | 0 | 0 | 0 | 0 |
Tetramer Bound States in Heteronuclear Systems | We calculate the universal spectrum of trimer and tetramer states in
heteronuclear mixtures of ultracold atoms with different masses in the vicinity
of the heavy-light dimer threshold. To extract the energies, we solve the
three- and four-body problem for simple two- and three-body potentials tuned to
the universal region using the Gaussian expansion method. We focus on the case
of one light particle of mass $m$ and two or three heavy bosons of mass $M$
with resonant heavy-light interactions. We find that trimer and tetramer cross
into the heavy-light dimer threshold at almost the same point and that as the
mass ratio $M/m$ decreases, the distance between the thresholds for trimer and
tetramer states becomes smaller. We also comment on the possibility of
observing exotic three-body states consisting of a dimer and two atoms in this
region and compare with previous work.
| 0 | 1 | 0 | 0 | 0 | 0 |
Dark Energy Cosmological Models with General forms of Scale Factor | In this paper, we have constructed dark energy models in an anisotropic
Bianchi-V space-time and studied the role of anisotropy in the evolution of
dark energy. We have considered anisotropic dark energy fluid with different
pressure gradients along different spatial directions. In order to obtain a
deterministic solution, we have considered three general forms of scale factor.
The different forms of scale factors considered here produce time varying
deceleration parameters in all the cases that simulates the cosmic transition.
The variable equation of state (EoS) parameter, skewness parameters for all the
models are obtained and analyzed. The physical properties of the models are
also discussed.
| 0 | 1 | 0 | 0 | 0 | 0 |
Mutual Kernel Matrix Completion | With the huge influx of various data nowadays, extracting knowledge from them
has become an interesting but tedious task among data scientists, particularly
when the data come in heterogeneous form and have missing information. Many
data completion techniques had been introduced, especially in the advent of
kernel methods. However, among the many data completion techniques available in
the literature, studies about mutually completing several incomplete kernel
matrices have not been given much attention yet. In this paper, we present a
new method, called Mutual Kernel Matrix Completion (MKMC) algorithm, that
tackles this problem of mutually inferring the missing entries of multiple
kernel matrices by combining the notions of data fusion and kernel matrix
completion, applied on biological data sets to be used for classification task.
We first introduced an objective function that will be minimized by exploiting
the EM algorithm, which in turn results to an estimate of the missing entries
of the kernel matrices involved. The completed kernel matrices are then
combined to produce a model matrix that can be used to further improve the
obtained estimates. An interesting result of our study is that the E-step and
the M-step are given in closed form, which makes our algorithm efficient in
terms of time and memory. After completion, the (completed) kernel matrices are
then used to train an SVM classifier to test how well the relationships among
the entries are preserved. Our empirical results show that the proposed
algorithm bested the traditional completion techniques in preserving the
relationships among the data points, and in accurately recovering the missing
kernel matrix entries. By far, MKMC offers a promising solution to the problem
of mutual estimation of a number of relevant incomplete kernel matrices.
| 1 | 0 | 0 | 1 | 0 | 0 |
Quantum Klein Space and Superspace | We give an algebraic quantization, in the sense of quantum groups, of the
complex Minkowski space, and we examine the real forms corresponding to the
signatures $(3,1)$, $(2,2)$, $(4,0)$, constructing the corresponding quantum
metrics and providing an explicit presentation of the quantized coordinate
algebras. In particular, we focus on the Kleinian signature $(2,2)$. The
quantizations of the complex and real spaces come together with a coaction of
the quantizations of the respective symmetry groups. We also extend such
quantizations to the $\mathcal{N}=1$ supersetting.
| 0 | 0 | 1 | 0 | 0 | 0 |
Bayesian Lasso Posterior Sampling via Parallelized Measure Transport | It is well known that the Lasso can be interpreted as a Bayesian posterior
mode estimate with a Laplacian prior. Obtaining samples from the full posterior
distribution, the Bayesian Lasso, confers major advantages in performance as
compared to having only the Lasso point estimate. Traditionally, the Bayesian
Lasso is implemented via Gibbs sampling methods which suffer from lack of
scalability, unknown convergence rates, and generation of samples that are
necessarily correlated. We provide a measure transport approach to generate
i.i.d samples from the posterior by constructing a transport map that
transforms a sample from the Laplacian prior into a sample from the posterior.
We show how the construction of this transport map can be parallelized into
modules that iteratively solve Lasso problems and perform closed-form linear
algebra updates. With this posterior sampling method, we perform maximum
likelihood estimation of the Lasso regularization parameter via the EM
algorithm. We provide comparisons to traditional Gibbs samplers using the
diabetes dataset of Efron et al. Lastly, we give an example implementation on a
computing system that leverages parallelization, a graphics processing unit,
whose execution time has much less dependence on dimension as compared to a
standard implementation.
| 0 | 0 | 0 | 1 | 0 | 0 |
Endogeneous Dynamics of Intraday Liquidity | In this paper we investigate the endogenous information contained in four
liquidity variables at a five minutes time scale on equity markets around the
world: the traded volume, the bid-ask spread, the volatility and the volume at
first limits of the orderbook. In the spirit of Granger causality, we measure
the level of information by the level of accuracy of linear autoregressive
models. This empirical study is carried out on a dataset of more than 300
stocks from four different markets (US, UK, Japan and Hong Kong) from a period
of over five years. We discuss the obtained performances of autoregressive (AR)
models on stationarized versions of the variables, focusing on explaining the
observed differences between stocks.
Since empirical studies are often conducted at this time scale, we believe it
is of paramount importance to document endogenous dynamics in a simple
framework with no addition of supplemental information. Our study can hence be
used as a benchmark to identify exogenous effects. On the other hand, most
optimal trading frameworks (like the celebrated Almgren and Chriss one), focus
on computing an optimal trading speed at a frequency close to the one we
consider. Such frameworks very often take i.i.d. assumptions on liquidity
variables; this paper document the auto-correlations emerging from real data,
opening the door to new developments in optimal trading.
| 0 | 0 | 0 | 0 | 0 | 1 |
Medical Image Synthesis for Data Augmentation and Anonymization using Generative Adversarial Networks | Data diversity is critical to success when training deep learning models.
Medical imaging data sets are often imbalanced as pathologic findings are
generally rare, which introduces significant challenges when training deep
learning models. In this work, we propose a method to generate synthetic
abnormal MRI images with brain tumors by training a generative adversarial
network using two publicly available data sets of brain MRI. We demonstrate two
unique benefits that the synthetic images provide. First, we illustrate
improved performance on tumor segmentation by leveraging the synthetic images
as a form of data augmentation. Second, we demonstrate the value of generative
models as an anonymization tool, achieving comparable tumor segmentation
results when trained on the synthetic data versus when trained on real subject
data. Together, these results offer a potential solution to two of the largest
challenges facing machine learning in medical imaging, namely the small
incidence of pathological findings, and the restrictions around sharing of
patient data.
| 0 | 0 | 0 | 1 | 0 | 0 |
Adaptive Feature Representation for Visual Tracking | Robust feature representation plays significant role in visual tracking.
However, it remains a challenging issue, since many factors may affect the
experimental performance. The existing method which combine different features
by setting them equally with the fixed weight could hardly solve the issues,
due to the different statistical properties of different features across
various of scenarios and attributes. In this paper, by exploiting the internal
relationship among these features, we develop a robust method to construct a
more stable feature representation. More specifically, we utilize a co-training
paradigm to formulate the intrinsic complementary information of multi-feature
template into the efficient correlation filter framework. We test our approach
on challenging se- quences with illumination variation, scale variation,
deformation etc. Experimental results demonstrate that the proposed method
outperforms state-of-the-art methods favorably.
| 1 | 0 | 0 | 0 | 0 | 0 |
An analysis of the SPARSEVA estimate for the finite sample data case | In this paper, we develop an upper bound for the SPARSEVA (SPARSe Estimation
based on a VAlidation criterion) estimation error in a general scheme, i.e.,
when the cost function is strongly convex and the regularized norm is
decomposable for a pair of subspaces. We show how this general bound can be
applied to a sparse regression problem to obtain an upper bound for the
traditional SPARSEVA problem. Numerical results are used to illustrate the
effectiveness of the suggested bound.
| 0 | 0 | 1 | 1 | 0 | 0 |
The Remarkable Similarity of Massive Galaxy Clusters From z~0 to z~1.9 | We present the results of a Chandra X-ray survey of the 8 most massive galaxy
clusters at z>1.2 in the South Pole Telescope 2500 deg^2 survey. We combine
this sample with previously-published Chandra observations of 49 massive
X-ray-selected clusters at 0<z<0.1 and 90 SZ-selected clusters at 0.25<z<1.2 to
constrain the evolution of the intracluster medium (ICM) over the past ~10 Gyr.
We find that the bulk of the ICM has evolved self similarly over the full
redshift range probed here, with the ICM density at r>0.2R500 scaling like
E(z)^2. In the centers of clusters (r<0.1R500), we find significant deviations
from self similarity (n_e ~ E(z)^{0.1+/-0.5}), consistent with no redshift
dependence. When we isolate clusters with over-dense cores (i.e., cool cores),
we find that the average over-density profile has not evolved with redshift --
that is, cool cores have not changed in size, density, or total mass over the
past ~9-10 Gyr. We show that the evolving "cuspiness" of clusters in the X-ray,
reported by several previous studies, can be understood in the context of a
cool core with fixed properties embedded in a self similarly-evolving cluster.
We find no measurable evolution in the X-ray morphology of massive clusters,
seemingly in tension with the rapidly-rising (with redshift) rate of major
mergers predicted by cosmological simulations. We show that these two results
can be brought into agreement if we assume that the relaxation time after a
merger is proportional to the crossing time, since the latter is proportional
to H(z)^(-1).
| 0 | 1 | 0 | 0 | 0 | 0 |
Rigorous estimates for the relegation algorithm | We revisit the relegation algorithm by Deprit et al. (Celest. Mech. Dyn.
Astron. 79:157-182, 2001) in the light of the rigorous Nekhoroshev's like
theory. This relatively recent algorithm is nowadays widely used for
implementing closed form analytic perturbation theories, as it generalises the
classical Birkhoff normalisation algorithm. The algorithm, here briefly
explained by means of Lie transformations, has been so far introduced and used
in a formal way, i.e. without providing any rigorous convergence or asymptotic
estimates. The overall aim of this paper is to find such quantitative estimates
and to show how the results about stability over exponentially long times can
be recovered in a simple and effective way, at least in the non-resonant case.
| 0 | 1 | 0 | 0 | 0 | 0 |
Linear Pentapods with a Simple Singularity Variety | There exists a bijection between the configuration space of a linear pentapod
and all points $(u,v,w,p_x,p_y,p_z)\in\mathbb{R}^{6}$ located on the singular
quadric $\Gamma: u^2+v^2+w^2=1$, where $(u,v,w)$ determines the orientation of
the linear platform and $(p_x,p_y,p_z)$ its position. Then the set of all
singular robot configurations is obtained by intersecting $\Gamma$ with a cubic
hypersurface $\Sigma$ in $\mathbb{R}^{6}$, which is only quadratic in the
orientation variables and position variables, respectively. This article
investigates the restrictions to be imposed on the design of this mechanism in
order to obtain a reduction in degree. In detail we study the cases where
$\Sigma$ is (1) linear in position variables, (2) linear in orientation
variables and (3) quadratic in total. The resulting designs of linear pentapods
have the advantage of considerably simplified computation of singularity-free
spheres in the configuration space. Finally we propose three kinematically
redundant designs of linear pentapods with a simple singularity surface.
| 1 | 0 | 0 | 0 | 0 | 0 |
Neural Networks as Interacting Particle Systems: Asymptotic Convexity of the Loss Landscape and Universal Scaling of the Approximation Error | Neural networks, a central tool in machine learning, have demonstrated
remarkable, high fidelity performance on image recognition and classification
tasks. These successes evince an ability to accurately represent high
dimensional functions, potentially of great use in computational and applied
mathematics. That said, there are few rigorous results about the representation
error and trainability of neural networks. Here we characterize both the error
and the scaling of the error with the size of the network by reinterpreting the
standard optimization algorithm used in machine learning applications,
stochastic gradient descent, as the evolution of a particle system with
interactions governed by a potential related to the objective or "loss"
function used to train the network. We show that, when the number $n$ of
parameters is large, the empirical distribution of the particles descends on a
convex landscape towards a minimizer at a rate independent of $n$. We establish
a Law of Large Numbers and a Central Limit Theorem for the empirical
distribution, which together show that the approximation error of the network
universally scales as $O(n^{-1})$. Remarkably, these properties do not depend
on the dimensionality of the domain of the function that we seek to represent.
Our analysis also quantifies the scale and nature of the noise introduced by
stochastic gradient descent and provides guidelines for the step size and batch
size to use when training a neural network. We illustrate our findings on
examples in which we train neural network to learn the energy function of the
continuous 3-spin model on the sphere. The approximation error scales as our
analysis predicts in as high a dimension as $d=25$.
| 0 | 0 | 0 | 1 | 0 | 0 |
Adaptive Similar Triangles Method: a Stable Alternative to Sinkhorn's Algorithm for Regularized Optimal Transport | In this paper, we are motivated by two important applications:
entropy-regularized optimal transport problem and road or IP traffic demand
matrix estimation by entropy model. Both of them include solving a special type
of optimization problem with linear equality constraints and objective given as
a sum of an entropy regularizer and a linear function. It is known that the
state-of-the-art solvers for this problem, which are based on Sinkhorn's method
(also known as RSA or balancing method), can fail to work, when the
entropy-regularization parameter is small. We consider the above optimization
problem as a particular instance of a general strongly convex optimization
problem with linear constraints. We propose a new algorithm to solve this
general class of problems. Our approach is based on the transition to the dual
problem. First, we introduce a new accelerated gradient method with adaptive
choice of gradient's Lipschitz constant. Then, we apply this method to the dual
problem and show, how to reconstruct an approximate solution to the primal
problem with provable convergence rate. We prove the rate $O(1/k^2)$, $k$ being
the iteration counter, both for the absolute value of the primal objective
residual and constraints infeasibility. Our method has similar to Sinkhorn's
method complexity of each iteration, but is faster and more stable numerically,
when the regularization parameter is small. We illustrate the advantage of our
method by numerical experiments for the two mentioned applications. We show
that there exists a threshold, such that, when the regularization parameter is
smaller than this threshold, our method outperforms the Sinkhorn's method in
terms of computation time.
| 0 | 0 | 1 | 0 | 0 | 0 |
Images of Ideals under Derivations and $\mathcal E$-Derivations of Univariate Polynomial Algebras over a Field of Characteristic Zero | Let $K$ be a field of characteristic zero and $x$ a free variable. A
$K$-$\mathcal E$-derivation of $K[x]$ is a $K$-linear map of the form
$\operatorname{I}-\phi$ for some $K$-algebra endomorphism $\phi$ of $K[x]$,
where $\operatorname{I}$ denotes the identity map of $K[x]$. In this paper we
study the image of an ideal of $K[x]$ under some $K$-derivations and
$K$-$\mathcal E$-derivations of $K[x]$. We show that the LFED conjecture
proposed in [Z4] holds for all $K$-$\mathcal E$-derivations and all locally
finite $K$-derivations of $K[x]$. We also show that the LNED conjecture
proposed in [Z4] holds for all locally nilpotent $K$-derivations of $K[x]$, and
also for all locally nilpotent $K$-$\mathcal E$-derivations of $K[x]$ and the
ideals $uK[x]$ such that either $u=0$, or $\operatorname{deg}\, u\le 1$, or $u$
has at least one repeated root in the algebraic closure of $K$. As a
bi-product, the homogeneous Mathieu subspaces (Mathieu-Zhao spaces) of the
univariate polynomial algebra over an arbitrary field have also been
classified.
| 0 | 0 | 1 | 0 | 0 | 0 |
Maximum genus of the Jenga like configurations | We treat the boundary of the union of blocks in the Jenga game as a surface
with a polyhedral structure and consider its genus. We generalize the game and
determine the maximum genus of the generalized game.
| 0 | 0 | 1 | 0 | 0 | 0 |
A Decidable Very Expressive Description Logic for Databases (Extended Version) | We introduce $\mathcal{DLR}^+$, an extension of the n-ary propositionally
closed description logic $\mathcal{DLR}$ to deal with attribute-labelled tuples
(generalising the positional notation), projections of relations, and global
and local objectification of relations, able to express inclusion, functional,
key, and external uniqueness dependencies. The logic is equipped with both TBox
and ABox axioms. We show how a simple syntactic restriction on the appearance
of projections sharing common attributes in a $\mathcal{DLR}^+$ knowledge base
makes reasoning in the language decidable with the same computational
complexity as $\mathcal{DLR}$. The obtained $\mathcal{DLR}^\pm$ n-ary
description logic is able to encode more thoroughly conceptual data models such
as EER, UML, and ORM.
| 1 | 0 | 0 | 0 | 0 | 0 |
Cloud-based Deep Learning of Big EEG Data for Epileptic Seizure Prediction | Developing a Brain-Computer Interface~(BCI) for seizure prediction can help
epileptic patients have a better quality of life. However, there are many
difficulties and challenges in developing such a system as a real-life support
for patients. Because of the nonstationary nature of EEG signals, normal and
seizure patterns vary across different patients. Thus, finding a group of
manually extracted features for the prediction task is not practical. Moreover,
when using implanted electrodes for brain recording massive amounts of data are
produced. This big data calls for the need for safe storage and high
computational resources for real-time processing. To address these challenges,
a cloud-based BCI system for the analysis of this big EEG data is presented.
First, a dimensionality-reduction technique is developed to increase
classification accuracy as well as to decrease the communication bandwidth and
computation time. Second, following a deep-learning approach, a stacked
autoencoder is trained in two steps for unsupervised feature extraction and
classification. Third, a cloud-computing solution is proposed for real-time
analysis of big EEG data. The results on a benchmark clinical dataset
illustrate the superiority of the proposed patient-specific BCI as an
alternative method and its expected usefulness in real-life support of epilepsy
patients.
| 1 | 0 | 0 | 1 | 0 | 0 |
Centrality measures for graphons: Accounting for uncertainty in networks | As relational datasets modeled as graphs keep increasing in size and their
data-acquisition is permeated by uncertainty, graph-based analysis techniques
can become computationally and conceptually challenging. In particular, node
centrality measures rely on the assumption that the graph is perfectly known --
a premise not necessarily fulfilled for large, uncertain networks. Accordingly,
centrality measures may fail to faithfully extract the importance of nodes in
the presence of uncertainty. To mitigate these problems, we suggest a
statistical approach based on graphon theory: we introduce formal definitions
of centrality measures for graphons and establish their connections to
classical graph centrality measures. A key advantage of this approach is that
centrality measures defined at the modeling level of graphons are inherently
robust to stochastic variations of specific graph realizations. Using the
theory of linear integral operators, we define degree, eigenvector, Katz and
PageRank centrality functions for graphons and establish concentration
inequalities demonstrating that graphon centrality functions arise naturally as
limits of their counterparts defined on sequences of graphs of increasing size.
The same concentration inequalities also provide high-probability bounds
between the graphon centrality functions and the centrality measures on any
sampled graph, thereby establishing a measure of uncertainty of the measured
centrality score. The same concentration inequalities also provide
high-probability bounds between the graphon centrality functions and the
centrality measures on any sampled graph, thereby establishing a measure of
uncertainty of the measured centrality score.
| 1 | 0 | 1 | 1 | 0 | 0 |
A time-periodic mechanical analog of the quantum harmonic oscillator | We theoretically investigate the stability and linear oscillatory behavior of
a naturally unstable particle whose potential energy is harmonically modulated.
We find this fundamental dynamical system is analogous in time to a quantum
harmonic oscillator. In a certain modulation limit, a.k.a. the Kapitza regime,
the modulated oscillator can behave like an effective classic harmonic
oscillator. But in the overlooked opposite limit, the stable modes of
vibrations are quantized in the modulation parameter space. By analogy with the
statistical interpretation of quantum physics, those modes can be characterized
by the time-energy uncertainty relation of a quantum harmonic oscillator.
Reducing the almost-periodic vibrational modes of the particle to their
periodic eigenfunctions, one can transform the original equation of motion to a
dimensionless Schrödinger stationary wave equation with a harmonic potential.
This reduction process introduces two features reminiscent of the quantum
realm: a wave-particle duality and a loss of causality that could legitimate a
statistical interpretation of the computed eigenfunctions. These results shed
new light on periodically time-varying linear dynamical systems and open an
original path in the recently revived field of quantum mechanical analogs.
| 0 | 1 | 0 | 0 | 0 | 0 |
Anharmonicity and the isotope effect in superconducting lithium at high pressures: a first-principles approach | Recent experiments [Schaeffer 2015] have shown that lithium presents an
extremely anomalous isotope effect in the 15-25 GPa pressure range. In this
article we have calculated the anharmonic phonon dispersion of $\mathrm{^7Li}$
and $\mathrm{^6Li}$ under pressure, their superconducting transition
temperatures, and the associated isotope effect. We have found a huge
anharmonic renormalization of a transverse acoustic soft mode along $\Gamma$K
in the fcc phase, the expected structure at the pressure range of interest. In
fact, the anharmonic correction dynamically stabilizes the fcc phase above 25
GPa. However, we have not found any anomalous scaling of the superconducting
temperature with the isotopic mass. Additionally, we have also analyzed whether
the two lithium isotopes adopting different structures could explain the
observed anomalous behavior. According to our enthalpy calculations including
zero-point motion and anharmonicity it would not be possible in a stable
regime.
| 0 | 1 | 0 | 0 | 0 | 0 |
Time-resolved polarimetry of the superluminous SN 2015bn with the Nordic Optical Telescope | We present imaging polarimetry of the superluminous supernova SN 2015bn,
obtained over nine epochs between $-$20 and $+$46 days with the Nordic Optical
Telescope. This was a nearby, slowly-evolving Type I superluminous supernova
that has been studied extensively and for which two epochs of
spectropolarimetry are also available. Based on field stars, we determine the
interstellar polarisation in the Galaxy to be negligible. The polarisation of
SN 2015bn shows a statistically significant increase during the last epochs,
confirming previous findings. Our well-sampled imaging polarimetry series
allows us to determine that this increase (from $\sim 0.54\%$ to $\gtrsim
1.10\%$) coincides in time with rapid changes that took place in the optical
spectrum. We conclude that the supernova underwent a `phase transition' at
around $+$20 days, when the photospheric emission shifted from an outer layer,
dominated by natal C and O, to a more aspherical inner core, dominated by
freshly nucleosynthesized material. This two-layered model might account for
the characteristic appearance and properties of Type I superluminous
supernovae.
| 0 | 1 | 0 | 0 | 0 | 0 |
Deep Bayesian Active Learning with Image Data | Even though active learning forms an important pillar of machine learning,
deep learning tools are not prevalent within it. Deep learning poses several
difficulties when used in an active learning setting. First, active learning
(AL) methods generally rely on being able to learn and update models from small
amounts of data. Recent advances in deep learning, on the other hand, are
notorious for their dependence on large amounts of data. Second, many AL
acquisition functions rely on model uncertainty, yet deep learning methods
rarely represent such model uncertainty. In this paper we combine recent
advances in Bayesian deep learning into the active learning framework in a
practical way. We develop an active learning framework for high dimensional
data, a task which has been extremely challenging so far, with very sparse
existing literature. Taking advantage of specialised models such as Bayesian
convolutional neural networks, we demonstrate our active learning techniques
with image data, obtaining a significant improvement on existing active
learning approaches. We demonstrate this on both the MNIST dataset, as well as
for skin cancer diagnosis from lesion images (ISIC2016 task).
| 1 | 0 | 0 | 1 | 0 | 0 |
Robust Optical Flow Estimation in Rainy Scenes | Optical flow estimation in the rainy scenes is challenging due to background
degradation introduced by rain streaks and rain accumulation effects in the
scene. Rain accumulation effect refers to poor visibility of remote objects due
to the intense rainfall. Most existing optical flow methods are erroneous when
applied to rain sequences because the conventional brightness constancy
constraint (BCC) and gradient constancy constraint (GCC) generally break down
in this situation. Based on the observation that the RGB color channels receive
raindrop radiance equally, we introduce a residue channel as a new data
constraint to reduce the effect of rain streaks. To handle rain accumulation,
our method decomposes the image into a piecewise-smooth background layer and a
high-frequency detail layer. It also enforces the BCC on the background layer
only. Results on both synthetic dataset and real images show that our algorithm
outperforms existing methods on different types of rain sequences. To our
knowledge, this is the first optical flow method specifically dealing with
rain.
| 1 | 0 | 0 | 0 | 0 | 0 |
Thermophysical Phenomena in Metal Additive Manufacturing by Selective Laser Melting: Fundamentals, Modeling, Simulation and Experimentation | Among the many additive manufacturing (AM) processes for metallic materials,
selective laser melting (SLM) is arguably the most versatile in terms of its
potential to realize complex geometries along with tailored microstructure.
However, the complexity of the SLM process, and the need for predictive
relation of powder and process parameters to the part properties, demands
further development of computational and experimental methods. This review
addresses the fundamental physical phenomena of SLM, with a special emphasis on
the associated thermal behavior. Simulation and experimental methods are
discussed according to three primary categories. First, macroscopic approaches
aim to answer questions at the component level and consider for example the
determination of residual stresses or dimensional distortion effects prevalent
in SLM. Second, mesoscopic approaches focus on the detection of defects such as
excessive surface roughness, residual porosity or inclusions that occur at the
mesoscopic length scale of individual powder particles. Third, microscopic
approaches investigate the metallurgical microstructure evolution resulting
from the high temperature gradients and extreme heating and cooling rates
induced by the SLM process. Consideration of physical phenomena on all of these
three length scales is mandatory to establish the understanding needed to
realize high part quality in many applications, and to fully exploit the
potential of SLM and related metal AM processes.
| 1 | 1 | 0 | 0 | 0 | 0 |
Numerical Methods for Pulmonary Image Registration | Due to complexity and invisibility of human organs, diagnosticians need to
analyze medical images to determine where the lesion region is, and which kind
of disease is, in order to make precise diagnoses. For satisfying clinical
purposes through analyzing medical images, registration plays an essential
role. For instance, in Image-Guided Interventions (IGI) and computer-aided
surgeries, patient anatomy is registered to preoperative images to guide
surgeons complete procedures. Medical image registration is also very useful in
surgical planning, monitoring disease progression and for atlas construction.
Due to the significance, the theories, methods, and implementation method of
image registration constitute fundamental knowledge in educational training for
medical specialists. In this chapter, we focus on image registration of a
specific human organ, i.e. the lung, which is prone to be lesioned. For
pulmonary image registration, the improvement of the accuracy and how to obtain
it in order to achieve clinical purposes represents an important problem which
should seriously be addressed. In this chapter, we provide a survey which
focuses on the role of image registration in educational training together with
the state-of-the-art of pulmonary image registration. In the first part, we
describe clinical applications of image registration introducing artificial
organs in Simulation-based Education. In the second part, we summarize the
common methods used in pulmonary image registration and analyze popular papers
to obtain a survey of pulmonary image registration.
| 0 | 1 | 0 | 0 | 0 | 0 |
Solving Non-parametric Inverse Problem in Continuous Markov Random Field using Loopy Belief Propagation | In this paper, we address the inverse problem, or the statistical machine
learning problem, in Markov random fields with a non-parametric pair-wise
energy function with continuous variables. The inverse problem is formulated by
maximum likelihood estimation. The exact treatment of maximum likelihood
estimation is intractable because of two problems: (1) it includes the
evaluation of the partition function and (2) it is formulated in the form of
functional optimization. We avoid Problem (1) by using Bethe approximation.
Bethe approximation is an approximation technique equivalent to the loopy
belief propagation. Problem (2) can be solved by using orthonormal function
expansion. Orthonormal function expansion can reduce a functional optimization
problem to a function optimization problem. Our method can provide an analytic
form of the solution of the inverse problem within the framework of Bethe
approximation.
| 1 | 1 | 0 | 1 | 0 | 0 |
Topology Adaptive Graph Convolutional Networks | Spectral graph convolutional neural networks (CNNs) require approximation to
the convolution to alleviate the computational complexity, resulting in
performance loss. This paper proposes the topology adaptive graph convolutional
network (TAGCN), a novel graph convolutional network defined in the vertex
domain. We provide a systematic way to design a set of fixed-size learnable
filters to perform convolutions on graphs. The topologies of these filters are
adaptive to the topology of the graph when they scan the graph to perform
convolution. The TAGCN not only inherits the properties of convolutions in CNN
for grid-structured data, but it is also consistent with convolution as defined
in graph signal processing. Since no approximation to the convolution is
needed, TAGCN exhibits better performance than existing spectral CNNs on a
number of data sets and is also computationally simpler than other recent
methods.
| 1 | 0 | 0 | 1 | 0 | 0 |
Suspensions of finite-size neutrally-buoyant spheres in turbulent duct flow | We study the turbulent square duct flow of dense suspensions of
neutrally-buoyant spherical particles. Direct numerical simulations (DNS) are
performed in the range of volume fractions $\phi=0-0.2$, using the immersed
boundary method (IBM) to account for the dispersed phase. Based on the
hydraulic diameter a Reynolds number of $5600$ is considered. We report flow
features and particle statistics specific to this geometry, and compare the
results to the case of two-dimensional channel flows. In particular, we observe
that for $\phi=0.05$ and $0.1$, particles preferentially accumulate on the
corner bisectors, close to the duct corners as also observed for laminar square
duct flows of same duct-to-particle size ratios. At the highest volume
fraction, particles preferentially accumulate in the core region. For channel
flows, in the absence of lateral confinement particles are found instead to be
uniformily distributed across the channel. We also observe that the intensity
of the cross-stream secondary flows increases (with respect to the unladen
case) with the volume fraction up to $\phi=0.1$, as a consequence of the high
concentration of particles along the corner bisector. For $\phi=0.2$ the
turbulence activity is strongly reduced and the intensity of the secondary
flows reduces below that of the unladen case. The friction Reynolds number
increases with $\phi$ in dilute conditions, as observed for channel flows.
However, for $\phi=0.2$ the mean friction Reynolds number decreases below the
value for $\phi=0.1$.
| 0 | 1 | 0 | 0 | 0 | 0 |
Far-from-equilibrium transport of excited carriers in nanostructures | Transport of charged carriers in regimes of strong non-equilibrium is
critical in a wide array of applications ranging from solar energy conversion
and semiconductor devices to quantum information. Plasmonic hot-carrier science
brings this regime of transport physics to the forefront since photo-excited
carriers must be extracted far from equilibrium to harvest their energy
efficiently. Here, we present a theoretical and computational framework,
Non-Equilibrium Scattering in Space and Energy (NESSE), to predict the spatial
evolution of carrier energy distributions that combines the best features of
phase-space (Boltzmann) and particle-based (Monte Carlo) methods. Within the
NESSE framework, we bridge first-principles electronic structure predictions of
plasmon decay and carrier collision integrals at the atomic scale, with
electromagnetic field simulations at the nano- to mesoscale. Finally, we apply
NESSE to predict spatially-resolved energy distributions of photo-excited
carriers that impact the surface of experimentally realizable plasmonic
nanostructures, enabling first-principles design of hot carrier devices.
| 0 | 1 | 0 | 0 | 0 | 0 |
On annihilators of bounded $(\frak g, \frak k)$-modules | Let $\frak g$ be a semisimple Lie algebra and $\frak k\subset\frak g$ be a
reductive subalgebra. We say that a $\frak g$-module $M$ is a bounded $(\frak
g, \frak k)$-module if $M$ is a direct sum of simple finite-dimensional $\frak
k$-modules and the multiplicities of all simple $\frak k$-modules in that
direct sum are universally bounded.
The goal of this article is to show that the "boundedness" property for a
simple $(\frak g, \frak k)$-module $M$ is equivalent to a property of the
associated variety of the annihilator of $M$ (this is the closure of a
nilpotent coadjoint orbit inside $\frak g^*$) under the assumption that the
main field is algebraically closed and of characteristic 0. In particular this
implies that if $M_1, M_2$ are simple $(\frak g, \frak k)$-modules such that
$M_1$ is bounded and the associated varieties of the annihilators of $M_1$ and
$M_2$ coincide then $M_2$ is also bounded. This statement is a geometric
analogue of a purely algebraic fact due to I. Penkov and V. Serganova and it
was posed as a conjecture in my Ph.D. thesis.
| 0 | 0 | 1 | 0 | 0 | 0 |
Neutron Star Planets: Atmospheric processes and habitability | Of the roughly 3000 neutron stars known, only a handful have sub-stellar
companions. The most famous of these are the low-mass planets around the
millisecond pulsar B1257+12. New evidence indicates that observational biases
could still hide a wide variety of planetary systems around most neutron stars.
We consider the environment and physical processes relevant to neutron star
planets, in particular the effect of X-ray irradiation and the relativistic
pulsar wind on the planetary atmosphere. We discuss the survival time of planet
atmospheres and the planetary surface conditions around different classes of
neutron stars, and define a neutron star habitable zone. Depending on as-yet
poorly constrained aspects of the pulsar wind, both Super-Earths around
B1257+12 could lie within its habitable zone.
| 0 | 1 | 0 | 0 | 0 | 0 |
Discrete Time-Crystalline Order in Cavity and Circuit QED Systems | Discrete time crystals are a recently proposed and experimentally observed
out-of-equilibrium dynamical phase of Floquet systems, where the stroboscopic
evolution of a local observable repeats itself at an integer multiple of the
driving period. We address this issue in a driven-dissipative setup, focusing
on the modulated open Dicke model, which can be implemented by cavity or
circuit QED systems. In the thermodynamic limit, we employ semiclassical
approaches and find rich dynamical phases on top of the discrete
time-crystalline order. In a deep quantum regime with few qubits, we find clear
signatures of a transient discrete time-crystalline behavior, which is absent
in the isolated counterpart. We establish a phenomenology of dissipative
discrete time crystals by generalizing the Landau theory of phase transitions
to Floquet open systems.
| 0 | 1 | 0 | 0 | 0 | 0 |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.