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 wave propagation analysis and supratransmission prediction of a metastable modular metastructure for adaptive non-reciprocal energy transmission | In this research, we investigate the nonlinear energy transmission phenomenon
in a reconfigurable and adaptable metastable modular metastructure. Numerical
studies on a 1D metastable chain uncover that when the driving frequency is
within the stopband of the periodic structure, there exists a threshold input
amplitude, beyond which sudden increase in the energy transmission can be
observed. This onset of transmission is due to nonlinear instability and is
known as supratransmission. We show that due to spatial asymmetry of
strategically configured constituents, such transmission thresholds could shift
considerably when the structure is excited from different ends and therefore
enabling the non-reciprocal energy transmission. We discover that the critical
threshold amplitude can be predicted analytically using a localized
nonlinear-linear model combining harmonic balancing and transfer matrix
analyses. Additionally, influences of important parameters on the change of
threshold amplitude are investigated to provide insight on synthesizing systems
with desired non-reciprocal characteristics. These investigations elucidate the
rich and intricate dynamics achievable by nonlinearity, asymmetry, and
metastability, and provide new insights and opportunities to accomplish
adaptable non-reciprocal wave energy transmission.
| 0 | 1 | 0 | 0 | 0 | 0 |
Advances in Variational Inference | Many modern unsupervised or semi-supervised machine learning algorithms rely
on Bayesian probabilistic models. These models are usually intractable and thus
require approximate inference. Variational inference (VI) lets us approximate a
high-dimensional Bayesian posterior with a simpler variational distribution by
solving an optimization problem. This approach has been successfully used in
various models and large-scale applications. In this review, we give an
overview of recent trends in variational inference. We first introduce standard
mean field variational inference, then review recent advances focusing on the
following aspects: (a) scalable VI, which includes stochastic approximations,
(b) generic VI, which extends the applicability of VI to a large class of
otherwise intractable models, such as non-conjugate models, (c) accurate VI,
which includes variational models beyond the mean field approximation or with
atypical divergences, and (d) amortized VI, which implements the inference over
local latent variables with inference networks. Finally, we provide a summary
of promising future research directions.
| 1 | 0 | 0 | 1 | 0 | 0 |
Security Against Impersonation Attacks in Distributed Systems | In a multi-agent system, transitioning from a centralized to a distributed
decision-making strategy can introduce vulnerability to adversarial
manipulation. We study the potential for adversarial manipulation in a class of
graphical coordination games where the adversary can pose as a friendly agent
in the game, thereby influencing the decision-making rules of a subset of
agents. The adversary's influence can cascade throughout the system, indirectly
influencing other agents' behavior and significantly impacting the emergent
collective behavior. The main results in this paper focus on characterizing
conditions under which the adversary's local influence can dramatically impact
the emergent global behavior, e.g., destabilize efficient Nash equilibria.
| 1 | 0 | 0 | 0 | 0 | 0 |
Large-Batch Training for LSTM and Beyond | Large-batch training approaches have enabled researchers to utilize
large-scale distributed processing and greatly accelerate deep-neural net (DNN)
training. For example, by scaling the batch size from 256 to 32K, researchers
have been able to reduce the training time of ResNet50 on ImageNet from 29
hours to 2.2 minutes (Ying et al., 2018). In this paper, we propose a new
approach called linear-epoch gradual-warmup (LEGW) for better large-batch
training. With LEGW, we are able to conduct large-batch training for both CNNs
and RNNs with the Sqrt Scaling scheme. LEGW enables Sqrt Scaling scheme to be
useful in practice and as a result we achieve much better results than the
Linear Scaling learning rate scheme. For LSTM applications, we are able to
scale the batch size by a factor of 64 without losing accuracy and without
tuning the hyper-parameters. For CNN applications, LEGW is able to achieve the
same accuracy even as we scale the batch size to 32K. LEGW works better than
previous large-batch auto-tuning techniques. LEGW achieves a 5.3X average
speedup over the baselines for four LSTM-based applications on the same
hardware. We also provide some theoretical explanations for LEGW.
| 1 | 0 | 0 | 1 | 0 | 0 |
One-shot and few-shot learning of word embeddings | Standard deep learning systems require thousands or millions of examples to
learn a concept, and cannot integrate new concepts easily. By contrast, humans
have an incredible ability to do one-shot or few-shot learning. For instance,
from just hearing a word used in a sentence, humans can infer a great deal
about it, by leveraging what the syntax and semantics of the surrounding words
tells us. Here, we draw inspiration from this to highlight a simple technique
by which deep recurrent networks can similarly exploit their prior knowledge to
learn a useful representation for a new word from little data. This could make
natural language processing systems much more flexible, by allowing them to
learn continually from the new words they encounter.
| 1 | 0 | 0 | 1 | 0 | 0 |
Exactly Robust Kernel Principal Component Analysis | We propose a novel method called robust kernel principal component analysis
(RKPCA) to decompose a partially corrupted matrix as a sparse matrix plus a
high or full-rank matrix whose columns are drawn from a nonlinear
low-dimensional latent variable model. RKPCA can be applied to many problems
such as noise removal and subspace clustering and is so far the only
unsupervised nonlinear method robust to sparse noises. We also provide
theoretical guarantees for RKPCA. The optimization of RKPCA is challenging
because it involves nonconvex and indifferentiable problems simultaneously. We
propose two nonconvex optimization algorithms for RKPCA: alternating direction
method of multipliers with backtracking line search and proximal linearized
minimization with adaptive step size. Comparative studies on synthetic data and
nature images corroborate the effectiveness and superiority of RKPCA in noise
removal and robust subspace clustering.
| 0 | 0 | 0 | 1 | 0 | 0 |
Moderate deviation analysis for classical communication over quantum channels | We analyse families of codes for classical data transmission over quantum
channels that have both a vanishing probability of error and a code rate
approaching capacity as the code length increases. To characterise the
fundamental tradeoff between decoding error, code rate and code length for such
codes we introduce a quantum generalisation of the moderate deviation analysis
proposed by Altug and Wagner as well as Polyanskiy and Verdu. We derive such a
tradeoff for classical-quantum (as well as image-additive) channels in terms of
the channel capacity and the channel dispersion, giving further evidence that
the latter quantity characterises the necessary backoff from capacity when
transmitting finite blocks of classical data. To derive these results we also
study asymmetric binary quantum hypothesis testing in the moderate deviations
regime. Due to the central importance of the latter task, we expect that our
techniques will find further applications in the analysis of other quantum
information processing tasks.
| 1 | 0 | 1 | 0 | 0 | 0 |
Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks | We study the training process of Deep Neural Networks (DNNs) from the Fourier
analysis perspective. Our starting point is a Frequency Principle (F-Principle)
--- DNNs initialized with small parameters often fit target functions from low
to high frequencies --- which was first proposed by Xu et al. (2018) and
Rahaman et al. (2018) on synthetic datasets. In this work, we first show the
universality of the F-Principle by demonstrating this phenomenon on
high-dimensional benchmark datasets, such as MNIST and CIFAR10. Then, based on
experiments, we show that the F-Principle provides insight into both the
success and failure of DNNs in different types of problems. Based on the
F-Principle, we further propose that DNN can be adopted to accelerate the
convergence of low frequencies for scientific computing problems, in which most
of the conventional methods (e.g., Jacobi method) exhibit the opposite
convergence behavior --- faster convergence for higher frequencies. Finally, we
prove a theorem for DNNs of one hidden layer as a first step towards a
mathematical explanation of the F-Principle. Our work indicates that the
F-Principle with Fourier analysis is a promising approach to the study of DNNs
because it seems ubiquitous, applicable, and explainable.
| 1 | 0 | 0 | 1 | 0 | 0 |
Efficient, Safe, and Probably Approximately Complete Learning of Action Models | In this paper we explore the theoretical boundaries of planning in a setting
where no model of the agent's actions is given. Instead of an action model, a
set of successfully executed plans are given and the task is to generate a plan
that is safe, i.e., guaranteed to achieve the goal without failing. To this
end, we show how to learn a conservative model of the world in which actions
are guaranteed to be applicable. This conservative model is then given to an
off-the-shelf classical planner, resulting in a plan that is guaranteed to
achieve the goal. However, this reduction from a model-free planning to a
model-based planning is not complete: in some cases a plan will not be found
even when such exists. We analyze the relation between the number of observed
plans and the likelihood that our conservative approach will indeed fail to
solve a solvable problem. Our analysis show that the number of trajectories
needed scales gracefully.
| 1 | 0 | 0 | 0 | 0 | 0 |
Uniform cohomological expansion of uniformly quasiregular mappings | Let $f\colon M \to M$ be a uniformly quasiregular self-mapping of a compact,
connected, and oriented Riemannian $n$-manifold $M$ without boundary, $n\ge 2$.
We show that, for $k \in \{0,\ldots, n\}$, the induced homomorphism $f^* \colon
H^k(M;\mathbb{R}) \to H^k(M;\mathbb{R})$, where $H^k(M;\mathbb{R})$ is the
$k$:th singular cohomology of $M$, is complex diagonalizable and the
eigenvalues of $f^*$ have modulus $(\mathrm{deg}\ f)^{k/n}$. As an application,
we obtain a degree restriction for uniformly quasiregular self-mappings of
closed manifolds. In the proof of the main theorem, we use a Sobolev--de Rham
cohomology based on conformally invariant differential forms and an induced
push-forward operator.
| 0 | 0 | 1 | 0 | 0 | 0 |
The logic of pseudo-uninorms and their residua | Our method of density elimination is generalized to the non-commutative
substructural logic GpsUL*. Then the standard completeness of GpsUL* follows as
a lemma by virtue of previous work by Metcalfe and Montagna. This result shows
that GpsUL* is the logic of pseudo-uninorms and their residua and answered the
question posed by Prof. Metcalfe, Olivetti, Gabbay and Tsinakis.
| 0 | 0 | 1 | 0 | 0 | 0 |
Computation of Green's functions through algebraic decomposition of operators | In this article we use linear algebra to improve the computational time for
the obtaining of Green's functions of linear differential equations with
reflection (DER). This is achieved by decomposing both the `reduced' equation
(the ODE associated to a given DER) and the corresponding two-point boundary
conditions.
| 0 | 0 | 1 | 0 | 0 | 0 |
Methods for finding leader--follower equilibria with multiple followers | The concept of leader--follower (or Stackelberg) equilibrium plays a central
role in a number of real--world applications of game theory. While the case
with a single follower has been thoroughly investigated, results with multiple
followers are only sporadic and the problem of designing and evaluating
computationally tractable equilibrium-finding algorithms is still largely open.
In this work, we focus on the fundamental case where multiple followers play a
Nash equilibrium once the leader has committed to a strategy---as we
illustrate, the corresponding equilibrium finding problem can be easily shown
to be $\mathcal{FNP}$--hard and not in Poly--$\mathcal{APX}$ unless
$\mathcal{P} = \mathcal{NP}$ and therefore it is one among the hardest problems
to solve and approximate. We propose nonconvex mathematical programming
formulations and global optimization methods to find both exact and approximate
equilibria, as well as a heuristic black box algorithm. All the methods and
formulations that we introduce are thoroughly evaluated computationally.
| 1 | 0 | 0 | 0 | 0 | 0 |
Pressure Induced Superconductivity in the New Compound ScZrCo1-$δ$ | It is widely perceived that the correlation effect may play an important role
in several unconventional superconducting families, such as cuprate, iron-based
and heavy-fermion superconductors. The application of high pressure can tune
the ground state properties and balance the localization and itineracy of
electrons in correlated systems, which may trigger unconventional
superconductivity. Moreover, non-centrosymmetric structure may induce the spin
triplet pairing which is very rare in nature. Here, we report a new compound
ScZrCo1-${\delta}$ crystallizing in the Ti2Ni structure with the space group of
FD3-MS without a spatial inversion center. The resistivity of the material at
ambient pressure shows a bad metal and weak semiconducting behavior.
Furthermore, specific heat and magnetic susceptibility measurements yield a
rather large value of Wilson ratio ~4.47. Both suggest a ground state with
correlation effect. By applying pressure, the up-going behavior of resistivity
in lowering temperature at ambient pressure is suppressed and gradually it
becomes metallic. At a pressure of about 19.5 GPa superconductivity emerges. Up
to 36.05 GPa, a superconducting transition at about 3.6 K with a quite high
upper critical field is observed. Our discovery here provides a new platform
for investigating the relationship between correlation effect and
superconductivity.
| 0 | 1 | 0 | 0 | 0 | 0 |
Universal Adversarial Perturbations Against Semantic Image Segmentation | While deep learning is remarkably successful on perceptual tasks, it was also
shown to be vulnerable to adversarial perturbations of the input. These
perturbations denote noise added to the input that was generated specifically
to fool the system while being quasi-imperceptible for humans. More severely,
there even exist universal perturbations that are input-agnostic but fool the
network on the majority of inputs. While recent work has focused on image
classification, this work proposes attacks against semantic image segmentation:
we present an approach for generating (universal) adversarial perturbations
that make the network yield a desired target segmentation as output. We show
empirically that there exist barely perceptible universal noise patterns which
result in nearly the same predicted segmentation for arbitrary inputs.
Furthermore, we also show the existence of universal noise which removes a
target class (e.g., all pedestrians) from the segmentation while leaving the
segmentation mostly unchanged otherwise.
| 1 | 0 | 0 | 1 | 0 | 0 |
Marginal sequential Monte Carlo for doubly intractable models | Bayesian inference for models that have an intractable partition function is
known as a doubly intractable problem, where standard Monte Carlo methods are
not applicable. The past decade has seen the development of auxiliary variable
Monte Carlo techniques (M{\o}ller et al., 2006; Murray et al., 2006) for
tackling this problem; these approaches being members of the more general class
of pseudo-marginal, or exact-approximate, Monte Carlo algorithms (Andrieu and
Roberts, 2009), which make use of unbiased estimates of intractable posteriors.
Everitt et al. (2017) investigated the use of exact-approximate importance
sampling (IS) and sequential Monte Carlo (SMC) in doubly intractable problems,
but focussed only on SMC algorithms that used data-point tempering. This paper
describes SMC samplers that may use alternative sequences of distributions, and
describes ways in which likelihood estimates may be improved adaptively as the
algorithm progresses, building on ideas from Moores et al. (2015). This
approach is compared with a number of alternative algorithms for doubly
intractable problems, including approximate Bayesian computation (ABC), which
we show is closely related to the method of M{\o}ller et al. (2006).
| 1 | 0 | 0 | 1 | 0 | 0 |
The impact of neutral impurity concentration on charge drift mobility in n-type germanium | The impact of neutral impurity scattering of electrons on the charge drift
mobility in high purity n-type germanium crystals at 77 Kelvin is investigated.
We calculated the contributions from ionized impurity scattering, lattice
scattering, and neutral impurity scattering to the total charge drift mobility
using theoretical models. The experimental data such as charge carrier
concentration, mobility and resistivity are measured by Hall Effect system at
77 Kelvin. The neutral impurity concentration is derived from the Matthiessen's
rule using the measured Hall mobility and ionized impurity concentration. The
radial distribution of the neutral impurity concentration in the self-grown
crystals is determined. Consequently, we demonstrated that neutral impurity
scattering is a significant contribution to the charge drift mobility, which
has a dependence on the concentration of neutral impurities in high purity
n-type germanium crystal.
| 0 | 1 | 0 | 0 | 0 | 0 |
Calabi-Yau hypersurfaces and SU-bordism | Batyrev constructed a family of Calabi-Yau hypersurfaces dual to the first
Chern class in toric Fano varieties. Using this construction, we introduce a
family of Calabi-Yau manifolds whose SU-bordism classes generate the special
unitary bordism ring
$\varOmega^{SU}\otimes\mathbb{Z}[\frac{1}{2}]\cong\mathbb{Z}[\frac{1}{2}][y_{i}\colon
i\ge 2]$. We also describe explicit Calabi-Yau representatives for
multiplicative generators of the SU-bordism ring in low dimensions.
| 0 | 0 | 1 | 0 | 0 | 0 |
Distribution-Based Categorization of Classifier Transfer Learning | Transfer Learning (TL) aims to transfer knowledge acquired in one problem,
the source problem, onto another problem, the target problem, dispensing with
the bottom-up construction of the target model. Due to its relevance, TL has
gained significant interest in the Machine Learning community since it paves
the way to devise intelligent learning models that can easily be tailored to
many different applications. As it is natural in a fast evolving area, a wide
variety of TL methods, settings and nomenclature have been proposed so far.
However, a wide range of works have been reporting different names for the same
concepts. This concept and terminology mixture contribute however to obscure
the TL field, hindering its proper consideration. In this paper we present a
review of the literature on the majority of classification TL methods, and also
a distribution-based categorization of TL with a common nomenclature suitable
to classification problems. Under this perspective three main TL categories are
presented, discussed and illustrated with examples.
| 1 | 0 | 0 | 0 | 0 | 0 |
Alexander invariants of periodic virtual knots | We show that every periodic virtual knot can be realized as the closure of a
periodic virtual braid and use this to study the Alexander invariants of
periodic virtual knots. If $K$ is a $q$-periodic and almost classical knot, we
show that its quotient knot $K_*$ is also almost classical, and in the case
$q=p^r$ is a prime power, we establish an analogue of Murasugi's congruence
relating the Alexander polynomials of $K$ and $K_*$ over the integers modulo
$p$. This result is applied to the problem of determining the possible periods
of a virtual knot $K$. One consequence is that if $K$ is an almost classical
knot with a nontrivial Alexander polynomial, then it is $p$-periodic for only
finitely many primes $p$. Combined with parity and Manturov projection, our
methods provide conditions that a general virtual knot must satisfy in order to
be $q$-periodic.
| 0 | 0 | 1 | 0 | 0 | 0 |
Near-linear time approximation algorithms for optimal transport via Sinkhorn iteration | Computing optimal transport distances such as the earth mover's distance is a
fundamental problem in machine learning, statistics, and computer vision.
Despite the recent introduction of several algorithms with good empirical
performance, it is unknown whether general optimal transport distances can be
approximated in near-linear time. This paper demonstrates that this ambitious
goal is in fact achieved by Cuturi's Sinkhorn Distances. This result relies on
a new analysis of Sinkhorn iteration, which also directly suggests a new greedy
coordinate descent algorithm, Greenkhorn, with the same theoretical guarantees.
Numerical simulations illustrate that Greenkhorn significantly outperforms the
classical Sinkhorn algorithm in practice.
| 1 | 0 | 0 | 1 | 0 | 0 |
Mathematical model of gender bias and homophily in professional hierarchies | Women have become better represented in business, academia, and government
over time, yet a dearth of women at the highest levels of leadership remains.
Sociologists have attributed the leaky progression of women through
professional hierarchies to various cultural and psychological factors, such as
self-segregation and bias. Here, we present a minimal mathematical model that
reveals the relative role that bias and homophily (self-seeking) may play in
the ascension of women through professional hierarchies. Unlike previous
models, our novel model predicts that gender parity is not inevitable, and
deliberate intervention may be required to achieve gender balance in several
fields. To validate the model, we analyze a new database of gender
fractionation over time for 16 professional hierarchies. We quantify the degree
of homophily and bias in each professional hierarchy, and we propose specific
interventions to achieve gender parity more quickly.
| 1 | 0 | 0 | 0 | 0 | 0 |
Monotonicity of non-pluripolar products and complex Monge-Ampère equations with prescribed singularity | We establish the monotonicity property for the mass of non-pluripolar
products on compact Kahler manifolds, and we initiate the study of complex
Monge-Ampere type equations with prescribed singularity type. Using the
variational method of Berman-Boucksom-Guedj-Zeriahi we prove existence and
uniqueness of solutions with small unbounded locus. We give applications to
Kahler-Einstein metrics with prescribed singularity, and we show that the
log-concavity property holds for non-pluripolar products with small unbounded
locus.
| 0 | 0 | 1 | 0 | 0 | 0 |
Fast and high-quality tetrahedral mesh generation from neuroanatomical scans | Creating tetrahedral meshes with anatomically accurate surfaces is critically
important for a wide range of model-based neuroimaging modalities. However,
computationally efficient brain meshing algorithms and software are greatly
lacking. Here, we report a fully automated open-source software to rapidly
create high-quality tetrahedral meshes from brain segmentations. Built upon
various open-source meshing utilities, the proposed meshing workflow allows
robust generation of complex head and brain mesh models from multi-label
volumes, tissue probability maps, surface meshes and their combinations. The
quality of the complex tissue boundaries is preserved through a surface-based
approach, allowing fine-grained control over the sizes and quality of the mesh
elements through explicit user-defined meshing criteria. The proposed meshing
pipeline is highly versatile and compatible with many commonly used brain
analysis tools, including SPM, FSL, FreeSurfer, and BrainSuite. With this
mesh-generation pipeline, we demonstrate that one can generate 3D full-head
meshes that combine scalp, skull, cerebrospinal fluid, gray matter, white
matter, and air cavities with a typical processing time of less than 40
seconds. This approach can also incorporate highly detailed cortical and white
matter surface meshes derived from FSL and FreeSurfer with tissue segmentation
data. Finally, a high-quality brain atlas mesh library for different age
groups, ranging from infants to elderlies, was built to demonstrate the
robustness of the proposed workflow, as well as to serve as a common platform
for simulation-based brain studies. Our open-source meshing software
"brain2mesh" and the human brain atlas mesh library can be downloaded at
this http URL.
| 0 | 1 | 0 | 0 | 0 | 0 |
Curriculum Dropout | Dropout is a very effective way of regularizing neural networks.
Stochastically "dropping out" units with a certain probability discourages
over-specific co-adaptations of feature detectors, preventing overfitting and
improving network generalization. Besides, Dropout can be interpreted as an
approximate model aggregation technique, where an exponential number of smaller
networks are averaged in order to get a more powerful ensemble. In this paper,
we show that using a fixed dropout probability during training is a suboptimal
choice. We thus propose a time scheduling for the probability of retaining
neurons in the network. This induces an adaptive regularization scheme that
smoothly increases the difficulty of the optimization problem. This idea of
"starting easy" and adaptively increasing the difficulty of the learning
problem has its roots in curriculum learning and allows one to train better
models. Indeed, we prove that our optimization strategy implements a very
general curriculum scheme, by gradually adding noise to both the input and
intermediate feature representations within the network architecture.
Experiments on seven image classification datasets and different network
architectures show that our method, named Curriculum Dropout, frequently yields
to better generalization and, at worst, performs just as well as the standard
Dropout method.
| 1 | 0 | 0 | 1 | 0 | 0 |
Self-dual and logarithmic representations of the twisted Heisenberg--Virasoro algebra at level zero | This paper is a continuation of arXiv:1405.1707. We present certain new
applications and generalizations of the free field realization of the twisted
Heisenberg-Virasoro algebra ${\mathcal H}$ at level zero.
We find explicit formulas for singular vectors in certain Verma modules. A
free field realization of self-dual modules for ${\mathcal H}$ is presented by
combining a bosonic construction of Whittaker modules from arXiv:1409.5354 with
a construction of logarithmic modules for vertex algebras. As an application,
we prove that there exists a non-split self-extension of irreducible self-dual
module which is a logarithmic module of rank two.
We construct a large family of logarithmic modules containing different types
of highest weight modules as subquotients. We believe that these logarithmic
modules are related with projective covers of irreducible modules in a suitable
category of ${\mathcal H}$-modules.
| 0 | 0 | 1 | 0 | 0 | 0 |
Distance weighted discrimination of face images for gender classification | We illustrate the advantages of distance weighted discrimination for
classification and feature extraction in a High Dimension Low Sample Size
(HDLSS) situation. The HDLSS context is a gender classification problem of face
images in which the dimension of the data is several orders of magnitude larger
than the sample size. We compare distance weighted discrimination with Fisher's
linear discriminant, support vector machines, and principal component analysis
by exploring their classification interpretation through insightful
visuanimations and by examining the classifiers' discriminant errors. This
analysis enables us to make new contributions to the understanding of the
drivers of human discrimination between males and females.
| 1 | 0 | 0 | 1 | 0 | 0 |
Partition-based Unscented Kalman Filter for Reconfigurable Battery Pack State Estimation using an Electrochemical Model | Accurate state estimation of large-scale lithium-ion battery packs is
necessary for the advanced control of batteries, which could potentially
increase their lifetime through e.g. reconfiguration. To tackle this problem,
an enhanced reduced-order electrochemical model is used here. This model allows
considering a wider operating range and thermal coupling between cells, the
latter turning out to be significant. The resulting nonlinear model is
exploited for state estimation through unscented Kalman filters (UKF). A sensor
network composed of one sensor node per battery cell is deployed. Each sensor
node is equipped with a local UKF, which uses available local measurements
together with additional information coming from neighboring sensor nodes. Such
state estimation scheme gives rise to a partition-based unscented Kalman filter
(PUKF). The method is validated on data from a detailed simulator for a battery
pack comprised of six cells, with reconfiguration capabilities. The results
show that the distributed approach outperforms the centralized one in terms of
computation time at the expense of a very low increase of mean-square
estimation error.
| 1 | 0 | 0 | 0 | 0 | 0 |
Towards Principled Methods for Training Generative Adversarial Networks | The goal of this paper is not to introduce a single algorithm or method, but
to make theoretical steps towards fully understanding the training dynamics of
generative adversarial networks. In order to substantiate our theoretical
analysis, we perform targeted experiments to verify our assumptions, illustrate
our claims, and quantify the phenomena. This paper is divided into three
sections. The first section introduces the problem at hand. The second section
is dedicated to studying and proving rigorously the problems including
instability and saturation that arize when training generative adversarial
networks. The third section examines a practical and theoretically grounded
direction towards solving these problems, while introducing new tools to study
them.
| 1 | 0 | 0 | 1 | 0 | 0 |
Co-Clustering for Multitask Learning | This paper presents a new multitask learning framework that learns a shared
representation among the tasks, incorporating both task and feature clusters.
The jointly-induced clusters yield a shared latent subspace where task
relationships are learned more effectively and more generally than in
state-of-the-art multitask learning methods. The proposed general framework
enables the derivation of more specific or restricted state-of-the-art
multitask methods. The paper also proposes a highly-scalable multitask learning
algorithm, based on the new framework, using conjugate gradient descent and
generalized \textit{Sylvester equations}. Experimental results on synthetic and
benchmark datasets show that the proposed method systematically outperforms
several state-of-the-art multitask learning methods.
| 1 | 0 | 0 | 1 | 0 | 0 |
Avoiding Communication in Proximal Methods for Convex Optimization Problems | The fast iterative soft thresholding algorithm (FISTA) is used to solve
convex regularized optimization problems in machine learning. Distributed
implementations of the algorithm have become popular since they enable the
analysis of large datasets. However, existing formulations of FISTA communicate
data at every iteration which reduces its performance on modern distributed
architectures. The communication costs of FISTA, including bandwidth and
latency costs, is closely tied to the mathematical formulation of the
algorithm. This work reformulates FISTA to communicate data at every k
iterations and reduce data communication when operating on large data sets. We
formulate the algorithm for two different optimization methods on the Lasso
problem and show that the latency cost is reduced by a factor of k while
bandwidth and floating-point operation costs remain the same. The convergence
rates and stability properties of the reformulated algorithms are similar to
the standard formulations. The performance of communication-avoiding FISTA and
Proximal Newton methods is evaluated on 1 to 1024 nodes for multiple benchmarks
and demonstrate average speedups of 3-10x with scaling properties that
outperform the classical algorithms.
| 1 | 0 | 0 | 0 | 0 | 0 |
Stochastic Development Regression on Non-Linear Manifolds | We introduce a regression model for data on non-linear manifolds. The model
describes the relation between a set of manifold valued observations, such as
shapes of anatomical objects, and Euclidean explanatory variables. The approach
is based on stochastic development of Euclidean diffusion processes to the
manifold. Defining the data distribution as the transition distribution of the
mapped stochastic process, parameters of the model, the non-linear analogue of
design matrix and intercept, are found via maximum likelihood. The model is
intrinsically related to the geometry encoded in the connection of the
manifold. We propose an estimation procedure which applies the Laplace
approximation of the likelihood function. A simulation study of the performance
of the model is performed and the model is applied to a real dataset of Corpus
Callosum shapes.
| 1 | 0 | 0 | 0 | 0 | 0 |
Prediction of Kidney Function from Biopsy Images Using Convolutional Neural Networks | A Convolutional Neural Network was used to predict kidney function in
patients with chronic kidney disease from high-resolution digital pathology
scans of their kidney biopsies. Kidney biopsies were taken from participants of
the NEPTUNE study, a longitudinal cohort study whose goal is to set up
infrastructure for observing the evolution of 3 forms of idiopathic nephrotic
syndrome, including developing predictors for progression of kidney disease.
The knowledge of future kidney function is desirable as it can identify
high-risk patients and influence treatment decisions, reducing the likelihood
of irreversible kidney decline.
| 0 | 0 | 0 | 1 | 0 | 0 |
Weighing neutrinos in dynamical dark energy models | We briefly review the recent results of constraining neutrino mass in
dynamical dark energy models using cosmological observations and summarize the
findings. (i) In dynamical dark energy models, compared to $\Lambda$CDM, the
upper limit of $\sum m_\nu$ can become larger and can also become smaller. In
the cases of phantom and early phantom (i.e., the quintom evolving from $w<-1$
to $w>-1$), the constraint on $\sum m_\nu$ becomes looser; but in the cases of
quintessence and early quintessence (i.e., the quintom evolving from $w>-1$ to
$w<-1$), the constraint on $\sum m_\nu$ becomes tighter. (ii) In the
holographic dark energy (HDE) model, the tightest constraint on $\sum m_\nu$,
i.e., $\sum m_\nu<0.105$ eV, is obtained, which is almost equal to the lower
limit of $\sum m_\nu$ of IH case. Thus, it is of great interest to find that
the future neutrino oscillation experiments would potentially offer a possible
falsifying scheme for the HDE model. (iii) The mass splitting of neutrinos can
influence the cosmological fits. We find that the NH case fits the current
observations slightly better than the IH case, although the difference of
$\chi^2$ of the two cases is still not significant enough to definitely
distinguish the neutrino mass hierarchy.
| 0 | 1 | 0 | 0 | 0 | 0 |
Can supersymmetry emerge at a quantum critical point? | Supersymmetry plays an important role in superstring theory and particle
physics, but has never been observed in experiments. At certain quantum
critical points of condensed matter systems, the fermionic excitations are
gapless due to the special electronic structure whereas the bosonic order
parameter is automatically gapless, offering a promising platform to realize
emergent supersymmetry by tuning a single parameter. Here, we study under what
circumstances can supersymmetry emerge in a quantum critical system. We
demonstrate that the Yukawa-type coupling between the gapless fermion and boson
may induce a number of highly nonlocal self-interacting terms in the effective
field theory of the boson. Only when such terms do not exist or are irrelevant,
could supersymmetry have the chance to be dynamically generated at low
energies. This strong constraint provides an important guidance for the
exploration of emergent supersymmetry in various condensed matter systems, and
also should be carefully considered in the study of quantum critical behaviors.
| 0 | 1 | 0 | 0 | 0 | 0 |
Five-parameter potential box with inverse square singular boundaries | Using the Tridiagonal Representation Approach, we obtain solutions (energy
spectrum and corresponding wavefunctions) for a new five-parameter potential
box with inverse square singularity at the boundaries.
| 0 | 1 | 0 | 0 | 0 | 0 |
Succinctness in subsystems of the spatial mu-calculus | In this paper we systematically explore questions of succinctness in modal
logics employed in spatial reasoning. We show that the closure operator,
despite being less expressive, is exponentially more succinct than the
limit-point operator, and that the $\mu$-calculus is exponentially more
succinct than the equally-expressive tangled limit operator. These results hold
for any class of spaces containing at least one crowded metric space or
containing all spaces based on ordinals below $\omega^\omega$, with the usual
limit operator. We also show that these results continue to hold even if we
enrich the less succinct language with the universal modality.
| 0 | 0 | 1 | 0 | 0 | 0 |
Evidence for electronically-driven ferroelectricity in the family of strongly correlated dimerized BEDT-TTF molecular conductors | By applying measurements of the dielectric constants and relative length
changes to the dimerized molecular conductor
$\kappa$-(BEDT-TTF)$_2$Hg(SCN)$_2$Cl, we provide evidence for order-disorder
type electronic ferroelectricity which is driven by charge order within the
(BEDT-TTF)$_2$ dimers and stabilized by a coupling to the anions. According to
our density functional theory calculations, this material is characterized by a
moderate strength of dimerization. This system thus bridges the gap between
strongly dimerized materials, often approximated as dimer-Mott systems at 1/2
filling, and non- or weakly dimerized systems at 1/4 filling exhibiting charge
order. Our results indicate that intra-dimer charge degrees of freedom are of
particular importance in correlated $\kappa$-(BEDT-TTF)$_2$X salts and can
create novel states, such as electronically-driven multiferroicity or
charge-order-induced quasi-1D spin liquids.
| 0 | 1 | 0 | 0 | 0 | 0 |
Testing FLUKA on neutron activation of Si and Ge at nuclear research reactor using gamma spectroscopy | Samples of two characteristic semiconductor sensor materials, silicon and
germanium, have been irradiated with neutrons produced at the RP-10 Nuclear
Research Reactor at 4.5 MW. Their radionuclides photon spectra have been
measured with high resolution gamma spectroscopy, quantifying four
radioisotopes ($^{28}$Al, $^{29}$Al for Si and $^{75}$Ge and $^{77}$Ge for Ge).
We have compared the radionuclides production and their emission spectrum data
with Monte Carlo simulation results from FLUKA. Thus we have tested FLUKA's low
energy neutron library (ENDF/B-VIIR) and decay photon scoring with respect to
the activation of these semiconductors. We conclude that FLUKA is capable of
predicting relative photon peak amplitudes, with gamma intensities greater than
1%, of produced radionuclides with an average uncertainty of 13%. This work
allows us to estimate the corresponding systematic error on neutron activation
simulation studies of these sensor materials.
| 0 | 1 | 0 | 0 | 0 | 0 |
Probabilistic Search for Structured Data via Probabilistic Programming and Nonparametric Bayes | Databases are widespread, yet extracting relevant data can be difficult.
Without substantial domain knowledge, multivariate search queries often return
sparse or uninformative results. This paper introduces an approach for
searching structured data based on probabilistic programming and nonparametric
Bayes. Users specify queries in a probabilistic language that combines standard
SQL database search operators with an information theoretic ranking function
called predictive relevance. Predictive relevance can be calculated by a fast
sparse matrix algorithm based on posterior samples from CrossCat, a
nonparametric Bayesian model for high-dimensional, heterogeneously-typed data
tables. The result is a flexible search technique that applies to a broad class
of information retrieval problems, which we integrate into BayesDB, a
probabilistic programming platform for probabilistic data analysis. This paper
demonstrates applications to databases of US colleges, global macroeconomic
indicators of public health, and classic cars. We found that human evaluators
often prefer the results from probabilistic search to results from a standard
baseline.
| 1 | 0 | 0 | 1 | 0 | 0 |
Superposition of p-superharmonic functions | The Dominative $p$-Laplace Operator is introduced. This operator is a
relative to the $p$-Laplacian, but with the distinguishing property of being
sublinear. It explains the superposition principle in the $p$-Laplace Equation.
| 0 | 0 | 1 | 0 | 0 | 0 |
Ramsey properties and extending partial automorphisms for classes of finite structures | We show that every free amalgamation class of finite structures with
relations and (symmetric) partial functions is a Ramsey class when enriched by
a free linear ordering of vertices. This is a common strengthening of the
Nešetřil-Rödl Theorem and the second and third authors' Ramsey
theorem for finite models (that is, structures with both relations and
functions). We also find subclasses with the ordering property. For languages
with relational symbols and unary functions we also show the extension property
for partial automorphisms (EPPA) of free amalgamation classes. These general
results solve several conjectures and provide an easy Ramseyness test for many
classes of structures.
| 1 | 0 | 1 | 0 | 0 | 0 |
Estimation of the shape of the density contours of star-shaped distributions | Elliptically contoured distributions generalize the multivariate normal
distributions in such a way that the density generators need not be
exponential. However, as the name suggests, elliptically contoured
distributions remain to be restricted in that the similar density contours
ought to be elliptical. Kamiya, Takemura and Kuriki [Star-shaped distributions
and their generalizations, Journal of Statistical Planning and Inference 138
(2008), 3429--3447] proposed star-shaped distributions, for which the density
contours are allowed to be boundaries of arbitrary similar star-shaped sets. In
the present paper, we propose a nonparametric estimator of the shape of the
density contours of star-shaped distributions, and prove its strong consistency
with respect to the Hausdorff distance. We illustrate our estimator by
simulation.
| 0 | 0 | 1 | 1 | 0 | 0 |
Aggregated Pairwise Classification of Statistical Shapes | The classification of shapes is of great interest in diverse areas ranging
from medical imaging to computer vision and beyond. While many statistical
frameworks have been developed for the classification problem, most are
strongly tied to early formulations of the problem - with an object to be
classified described as a vector in a relatively low-dimensional Euclidean
space. Statistical shape data have two main properties that suggest a need for
a novel approach: (i) shapes are inherently infinite dimensional with strong
dependence among the positions of nearby points, and (ii) shape space is not
Euclidean, but is fundamentally curved. To accommodate these features of the
data, we work with the square-root velocity function of the curves to provide a
useful formal description of the shape, pass to tangent spaces of the manifold
of shapes at different projection points which effectively separate shapes for
pairwise classification in the training data, and use principal components
within these tangent spaces to reduce dimensionality. We illustrate the impact
of the projection point and choice of subspace on the misclassification rate
with a novel method of combining pairwise classifiers.
| 1 | 0 | 0 | 1 | 0 | 0 |
The MUSE-Wide survey: Detection of a clustering signal from Lyman-α-emitters at 3<z<6 | We present a clustering analysis of a sample of 238 Ly{$\alpha$}-emitters at
redshift 3<z<6 from the MUSE-Wide survey. This survey mosaics extragalactic
legacy fields with 1h MUSE pointings to detect statistically relevant samples
of emission line galaxies. We analysed the first year observations from
MUSE-Wide making use of the clustering signal in the line-of-sight direction.
This method relies on comparing pair-counts at close redshifts for a fixed
transverse distance and thus exploits the full potential of the redshift range
covered by our sample. A clear clustering signal with a correlation length of
r0 = 2.9(+1.0/-1.1) Mpc (comoving) is detected. Whilst this result is based on
only about a quarter of the full survey size, it already shows the immense
potential of MUSE for efficiently observing and studying the clustering of
Ly{$\alpha$}-emitters.
| 0 | 1 | 0 | 0 | 0 | 0 |
On Polymorphic Sessions and Functions: A Tale of Two (Fully Abstract) Encodings | This work exploits the logical foundation of session types to determine what
kind of type discipline for the pi-calculus can exactly capture, and is
captured by, lambda-calculus behaviours. Leveraging the proof theoretic content
of the soundness and completeness of sequent calculus and natural deduction
presentations of linear logic, we develop the first mutually inverse and fully
abstract processes-as-functions and functions-as-processes encodings between a
polymorphic session pi-calculus and a linear formulation of System F. We are
then able to derive results of the session calculus from the theory of the
lambda-calculus: (1) we obtain a characterisation of inductive and coinductive
session types via their algebraic representations in System F; and (2) we
extend our results to account for value and process passing, entailing strong
normalisation.
| 1 | 0 | 0 | 0 | 0 | 0 |
Effects of initial spatial phase in radiative neutrino pair emission | We study radiative neutrino pair emission in deexcitation process of atoms
taking into account coherence effect in a macroscopic target system. In the
course of preparing the coherent initial state to enhance the rate, a spatial
phase factor is imprinted in the macroscopic target. It is shown that this
initial spatial phase changes the kinematics of the radiative neutrino pair
emission. We investigate effects of the initial spatial phase in the photon
spectrum of the process. It turns out that the initial spatial phase provides
us significant improvements in exploring neutrino physics such as the
Dirac-Majorana distinction and the cosmic neutrino background.
| 0 | 1 | 0 | 0 | 0 | 0 |
Finite-sample risk bounds for maximum likelihood estimation with arbitrary penalties | The MDL two-part coding $ \textit{index of resolvability} $ provides a
finite-sample upper bound on the statistical risk of penalized likelihood
estimators over countable models. However, the bound does not apply to
unpenalized maximum likelihood estimation or procedures with exceedingly small
penalties. In this paper, we point out a more general inequality that holds for
arbitrary penalties. In addition, this approach makes it possible to derive
exact risk bounds of order $1/n$ for iid parametric models, which improves on
the order $(\log n)/n$ resolvability bounds. We conclude by discussing
implications for adaptive estimation.
| 0 | 0 | 1 | 1 | 0 | 0 |
Integrated Modeling of Second Phase Precipitation in Cold-Worked 316 Stainless Steels under Irradiation | The current work combines the Cluster Dynamics (CD) technique and
CALPHAD-based precipitation modeling to address the second phase precipitation
in cold-worked (CW) 316 stainless steels (SS) under irradiation at 300-400 C.
CD provides the radiation enhanced diffusion and dislocation evolution as
inputs for the precipitation model. The CALPHAD-based precipitation model
treats the nucleation, growth and coarsening of precipitation processes based
on classical nucleation theory and evolution equations, and simulates the
composition, size and size distribution of precipitate phases. We benchmark the
model against available experimental data at fast reactor conditions (9.4 x
10^-7 dpa/s and 390 C) and then use the model to predict the phase instability
of CW 316 SS under light water reactor (LWR) extended life conditions (7 x
10^-8 dpa/s and 275 C). The model accurately predicts the gamma-prime (Ni3Si)
precipitation evolution under fast reactor conditions and that the formation of
this phase is dominated by radiation enhanced segregation. The model also
predicts a carbide volume fraction that agrees well with available experimental
data from a PWR reactor but is much higher than the volume fraction observed in
fast reactors. We propose that radiation enhanced dissolution and/or carbon
depletion at sinks that occurs at high flux could be the main sources of this
inconsistency. The integrated model predicts ~1.2% volume fraction for carbide
and ~3.0% volume fraction for gamma-prime for typical CW 316 SS (with 0.054
wt.% carbon) under LWR extended life conditions. This work provides valuable
insights into the magnitudes and mechanisms of precipitation in irradiated CW
316 SS for nuclear applications.
| 0 | 1 | 0 | 0 | 0 | 0 |
Emergent topological superconductivity at nematic domain wall of FeSe | One dimensional hybrid systems play an important role in the search for
topological superconductivity. Nevertheless, all one dimensional hybrid systems
so far have been externally defined. Here we show that one-dimensional domain
wall in a nematic superconductor can serve as an emergent hybrid system in the
presence of spin-orbit coupling. As a concrete setting we study the domain wall
between nematic domains in FeSe, which is well established to be a nematic
superconductor. We first show on the symmetry grounds that spin-triplet pairing
can be induced at the domain wall by constructing a Ginzburg-Landau theory. We
then demonstrate using Bogoliubov-de Gennes approach that such nematic domain
wall supports zero energy bound states which would satisfy Majorana condition.
Well-known existence of these domain walls at relatively high temperatures,
which can in principle be located and investigated with scanning tunneling
microscopy, presents new opportunities for a search for realization of Majorana
bound states.
| 0 | 1 | 0 | 0 | 0 | 0 |
Linear and nonlinear market correlations: characterizing financial crises and portfolio optimization | Pearson correlation and mutual information based complex networks of the
day-to-day returns of US S&P500 stocks between 1985 and 2015 have been
constructed in order to investigate the mutual dependencies of the stocks and
their nature. We show that both networks detect qualitative differences
especially during (recent) turbulent market periods thus indicating strongly
fluctuating interconnections between the stocks of different companies in
changing economic environments. A measure for the strength of nonlinear
dependencies is derived using surrogate data and leads to interesting
observations during periods of financial market crises. In contrast to the
expectation that dependencies reduce mainly to linear correlations during
crises we show that (at least in the 2008 crisis) nonlinear effects are
significantly increasing. It turns out that the concept of centrality within a
network could potentially be used as some kind of an early warning indicator
for abnormal market behavior as we demonstrate with the example of the 2008
subprime mortgage crisis. Finally, we apply a Markowitz mean variance portfolio
optimization and integrate the measure of nonlinear dependencies to scale the
investment exposure. This leads to significant outperformance as compared to a
fully invested portfolio.
| 0 | 1 | 0 | 0 | 0 | 0 |
An explicit projective bimodule resolution of a Leavitt path algebra | We construct an explicit projective bimodule resolution for the Leavitt path
algebra of a row-finite quiver. We prove that the Leavitt path algebra of a
row-countable quiver has Hochschild cohomolgical dimension at most one, that
is, it is quasi-free in the sense of Cuntz-Quillen. The construction of the
resolution relies on an explicit derivation of the Leavitt path algebra.
| 0 | 0 | 1 | 0 | 0 | 0 |
Error Analysis and Improving the Accuracy of Winograd Convolution for Deep Neural Networks | Modern deep neural networks (DNNs) spend a large amount of their execution
time computing convolutions. Winograd's minimal algorithm for small
convolutions can greatly reduce the number of arithmetic operations. However, a
large reduction in floating point (FP) operations in these algorithms can
result in poor numeric accuracy. In this paper we analyse the FP error and
prove boundaries on the error. We show that the "modified" algorithm gives a
significantly better accuracy of the result. We propose several methods for
reducing FP error of these algorithms. Minimal convolution algorithms depend on
the selection of several numeric \textit{points} that have a large impact on
the accuracy of the result. We propose a canonical evaluation ordering that
both reduces FP error and the size of the search space based on Huffman coding.
We study point selection experimentally, and find empirically good points. We
also identify the main factors that associated with sets of points that result
in a low error. In addition, we explore other methods to reduce FP error,
including mixed-precision convolution, and pairwise addition across DNN
channels. Using our methods we can significantly reduce FP error for a given
block size, which allows larger block sizes and reduced computation.
| 1 | 0 | 0 | 1 | 0 | 0 |
Active Learning amidst Logical Knowledge | Structured prediction is ubiquitous in applications of machine learning such
as knowledge extraction and natural language processing. Structure often can be
formulated in terms of logical constraints. We consider the question of how to
perform efficient active learning in the presence of logical constraints among
variables inferred by different classifiers. We propose several methods and
provide theoretical results that demonstrate the inappropriateness of employing
uncertainty guided sampling, a commonly used active learning method.
Furthermore, experiments on ten different datasets demonstrate that the methods
significantly outperform alternatives in practice. The results are of practical
significance in situations where labeled data is scarce.
| 1 | 0 | 0 | 0 | 0 | 0 |
On the Limiting Stokes' Wave of Extreme Height in Arbitrary Water Depth | As mentioned by Schwartz (1974) and Cokelet (1977), it was failed to gain
convergent results of limiting Stokes' waves in extremely shallow water by
means of perturbation methods even with the aid of extrapolation techniques
such as Padé approximant. Especially, it is extremely difficult for
traditional analytic/numerical approaches to present the wave profile of
limiting waves with a sharp crest of $120^\circ$ included angle first mentioned
by Stokes in 1880s. Thus, traditionally, different wave models are used for
waves in different water depths. In this paper, by means of the homotopy
analysis method (HAM), an analytic approximation method for highly nonlinear
equations, we successfully gain convergent results (and especially the wave
profiles) of the limiting Stokes' waves with this kind of sharp crest in
arbitrary water depth, even including solitary waves of extreme form in
extremely shallow water, without using any extrapolation techniques. Therefore,
in the frame of the HAM, the Stokes' wave can be used as a unified theory for
all kinds of waves, including periodic waves in deep and intermediate depth,
cnoidal waves in shallow water and solitary waves in extremely shallow water.
| 0 | 1 | 0 | 0 | 0 | 0 |
Pressure impact on the stability and distortion of the crystal structure of CeScO3 | The effects of high pressure on the crystal structure of orthorhombic (Pnma)
perovskite type cerium scandate have been studied in situ under high pressure
by means of synchrotron x-ray powder diffraction, using a diamond anvil cell.
We have found that the perovskite type crystal structure remains stable up to
40 GPa, the highest pressure reached in the experiments. The evolution of
unit-cell parameters with pressure has indicated an anisotropic compression.
The room-temperature pressure-volume equation of state is obtained from the
experiments. From the evolution of microscopic structural parameters like bond
distances and coordination polyhedra of cerium and scandium, the macroscopic
behavior of CeScO3 under compression has been explained and reasoned for its
large pressure stability. The reported results are discussed in comparison with
high-pressure results from other perovskites.
| 0 | 1 | 0 | 0 | 0 | 0 |
The system of cloud oriented learning tools as an element of educational and scientific environment of high school | The aim of this research is to design and implementation of cloud based
learning environment for separate division of the university. The analysis of
existing approaches to the construction of cloud based learning environments,
the formation of requirements cloud based learning tools, the selection on the
basis of these requirements, cloud ICT training and pilot their use for
building cloud based learning environment for separate division of the
university with the use of open source software and resources its own IT
infrastructure of the institution. Results of the study is planned to
generalize to develop recommendations for the design of cloud based environment
of high school.
| 1 | 0 | 0 | 0 | 0 | 0 |
Chemical abundances of two extragalactic young massive clusters | We use integrated-light spectroscopic observations to measure metallicities
and chemical abundances for two extragalactic young massive star clusters
(NGC1313-379 and NGC1705-1). The spectra were obtained with the X-Shooter
spectrograph on the ESO Very Large Telescope. We compute synthetic
integrated-light spectra, based on colour-magnitude diagrams for the brightest
stars in the clusters from Hubble Space Telescope photometry and theoretical
isochrones. Furthermore, we test the uncertainties arising from the use of
Colour Magnitude Diagram (CMD) +Isochrone method compared to an Isochrone-Only
method. The abundances of the model spectra are iteratively adjusted until the
best fit to the observations is obtained. In this work we mainly focus on the
optical part of the spectra. We find metallicities of [Fe/H] = $-$0.84 $\pm$
0.07 and [Fe/H] = $-$0.78 $\pm$ 0.10 for NGC1313-379 and NGC1705-1,
respectively. We measure [$\alpha$/Fe]=$+$0.06 $\pm$ 0.11 for NGC1313-379 and a
super-solar [$\alpha$/Fe]=$+$0.32 $\pm$ 0.12 for NGC1705-1. The roughly solar
[$\alpha$/Fe] ratio in NGC1313-379 resembles those for young stellar
populations in the Milky Way (MW) and the Magellanic Clouds, whereas the
enhanced [$\alpha$/Fe] ratio in NGC1705-1 is similar to that found for the
cluster NGC1569-B by previous studies. Such super-solar [$\alpha$/Fe] ratios
are also predicted by chemical evolution models that incorporate the bursty
star formation histories of these dwarf galaxies. Furthermore, our
$\alpha$-element abundances agree with abundance measurements from H II regions
in both galaxies. In general we derive Fe-peak abundances similar to those
observed in the MW and Large Magellanic Cloud (LMC) for both young massive
clusters. For these elements, however, we recommend higher-resolution
observations to improve the Fe-peak abundance measurements.
| 0 | 1 | 0 | 0 | 0 | 0 |
Clingo goes Linear Constraints over Reals and Integers | The recent series 5 of the ASP system clingo provides generic means to
enhance basic Answer Set Programming (ASP) with theory reasoning capabilities.
We instantiate this framework with different forms of linear constraints,
discuss the respective implementations, and present techniques of how to use
these constraints in a reactive context. More precisely, we introduce
extensions to clingo with difference and linear constraints over integers and
reals, respectively, and realize them in complementary ways. Finally, we
empirically evaluate the resulting clingo derivatives clingo[dl] and clingo[lp]
on common fragments and contrast them to related ASP systems.
This paper is under consideration for acceptance in TPLP.
| 1 | 0 | 0 | 0 | 0 | 0 |
Dynamic Uplink/Downlink Resource Management in Flexible Duplex-Enabled Wireless Networks | Flexible duplex is proposed to adapt to the channel and traffic asymmetry for
future wireless networks. In this paper, we propose two novel algorithms within
the flexible duplex framework for joint uplink and downlink resource allocation
in multi-cell scenario, named SAFP and RMDI, based on the awareness of
interference coupling among wireless links. Numerical results show significant
performance gain over the baseline system with fixed uplink/downlink resource
configuration, and over the dynamic TDD scheme that independently adapts the
configuration to time-varying traffic volume in each cell. The proposed
algorithms achieve two-fold increase when compared with the baseline scheme,
measured by the worst-case quality of service satisfaction level, under a low
level of traffic asymmetry. The gain is more significant when the traffic is
highly asymmetric, as it achieves three-fold increase.
| 1 | 0 | 0 | 0 | 0 | 0 |
Carleman estimates for forward and backward stochastic fourth order Schrödinger equations and their applications | In this paper, we establish the Carleman estimates for forward and backward
stochastic fourth order Schrödinger equations, on basis of which, we can
obtain the observability, unique continuation property and the exact
controllability for the forward and backward stochastic fourth order
Schrödinger equations.
| 0 | 0 | 1 | 0 | 0 | 0 |
Reply to comment on `Poynting flux in the neighbourhood of a point charge in arbitrary motion and the radiative power losses' | Doubts have been expressed in a comment (Eur. J. Phys., 39, 018001, 2018),
about the tenability of the formulation for radiative losses in our recent
published work (Eur. J. Phys., 37, 045210, 2016). We provide our reply to the
comment. In particular, it is pointed out that one need to clearly distinguish
between the rate of the energy-momentum being carried by the electromagnetic
radiation to far-off space, and that of the mechanical energy-momentum losses
being incurred by the radiating charge. It is also demonstrated that while the
Poynting flux is always positive through a spherical surface centred on the
retarded position of the charge, it could surprisingly be negative through a
surface centred on the "present" position of the charge. It is further shown
that the mysterious Schott term, hitherto thought in literature to arise from
some acceleration-dependent energy in fields, is actually nothing but the
difference in rate of change of energy in self-fields of the charge between the
retarded and present times.
| 0 | 1 | 0 | 0 | 0 | 0 |
Delegated Causality of Complex Systems | A notion of delegated causality is introduced. This subtle kind of causality
is dual to interventional causality. Delegated causality elucidates the causal
role of dynamical systems at the "edge of chaos", explicates evident cases of
downward causation, and relates emergent phenomena to Godel's incompleteness
theorem. Apparently rich implications are noticed in biology and Chinese
philosophy.
| 0 | 1 | 0 | 0 | 0 | 0 |
Resonating Valence Bond Theory of Superconductivity: Beyond Cuprates | Resonating valence bond (RVB) theory of high Tc superconductivity, an
electron correlation based mechanism, began as an insightful response by
Anderson, to Bednorz and Muller's discovery of high Tc superconductivity in
cuprates in late 1986. Shortly a theoretical framework for quantum spin liquids
and superconductivity was developed. This theory adresses a formidable strong
coupling quantum manybody problem, in modern times. It is built on certain key
experimental facts: i) survival of a dynamical Mott localization in a metallic
state, ii) proliferation of bond singlets and iii) absence of fermi liquid
quasi particles. After summarising RVB theory I will provide an aerial view of,
mostly, new superconductors where I believe that, to a large degree RVB
mechanism is at work and indicate prospects for even higher Tc's.
| 0 | 1 | 0 | 0 | 0 | 0 |
Efficient and accurate numerical schemes for a hydrodynamically coupled phase field diblock copolymer model | In this paper, we consider numerical approximations of a hydrodynamically
coupled phase field diblock copolymer model, in which the free energy contains
a kinetic potential, a gradient entropy, a Ginzburg-Landau double well
potential, and a long range nonlocal type potential. We develop a set of second
order time marching schemes for this system using the "Invariant Energy
Quadratization" approach for the double well potential, the projection method
for the Navier-Stokes equation, and a subtle implicit-explicit treatment for
the stress and convective term. The resulting schemes are linear and lead to
symmetric positive definite systems at each time step, thus they can be
efficiently solved. We further prove that these schemes are unconditionally
energy stable. Various numerical experiments are performed to validate the
accuracy and energy stability of the proposed schemes.
| 0 | 0 | 1 | 0 | 0 | 0 |
Brownian Motion of a Classical Particle in Quantum Environment | The Klein-Kramers equation, governing the Brownian motion of a classical
particle in quantum environment under the action of an arbitrary external
potential, is derived. Quantum temperature and friction operators are
introduced and at large friction the corresponding Smoluchowski equation is
obtained. Introducing the Bohm quantum potential, this Smoluchowski equation is
extended to describe the Brownian motion of a quantum particle in quantum
environment.
| 0 | 1 | 0 | 0 | 0 | 0 |
Hierarchical Policy Search via Return-Weighted Density Estimation | Learning an optimal policy from a multi-modal reward function is a
challenging problem in reinforcement learning (RL). Hierarchical RL (HRL)
tackles this problem by learning a hierarchical policy, where multiple option
policies are in charge of different strategies corresponding to modes of a
reward function and a gating policy selects the best option for a given
context. Although HRL has been demonstrated to be promising, current
state-of-the-art methods cannot still perform well in complex real-world
problems due to the difficulty of identifying modes of the reward function. In
this paper, we propose a novel method called hierarchical policy search via
return-weighted density estimation (HPSDE), which can efficiently identify the
modes through density estimation with return-weighted importance sampling. Our
proposed method finds option policies corresponding to the modes of the return
function and automatically determines the number and the location of option
policies, which significantly reduces the burden of hyper-parameters tuning.
Through experiments, we demonstrate that the proposed HPSDE successfully learns
option policies corresponding to modes of the return function and that it can
be successfully applied to a challenging motion planning problem of a redundant
robotic manipulator.
| 1 | 0 | 0 | 1 | 0 | 0 |
Two-dimensional Fourier transformations and Mordell integrals | Several Fourier transformations of functions of one and two variables are
evaluated and then used to derive some integral and series identities. It is
shown that certain two- dimensional Mordell integrals factorize into product of
two integrals and that the square of the absolute value of the Mordell integral
can be reduced to a single one-dimensional integral. Some connections to
elliptic functions and lattice sums are discussed.
| 0 | 0 | 1 | 0 | 0 | 0 |
Optimal Allocation of Static Var Compensator via Mixed Integer Conic Programming | Shunt FACTS devices, such as, a Static Var Compensator (SVC), are capable of
providing local reactive power compensation. They are widely used in the
network to reduce the real power loss and improve the voltage profile. This
paper proposes a planning model based on mixed integer conic programming (MICP)
to optimally allocate SVCs in the transmission network considering load
uncertainty. The load uncertainties are represented by a number of scenarios.
Reformulation and linearization techniques are utilized to transform the
original non-convex model into a convex second order cone programming (SOCP)
model. Numerical case studies based on the IEEE 30-bus system demonstrate the
effectiveness of the proposed planning model.
| 0 | 0 | 1 | 0 | 0 | 0 |
Polarity tuning of spin-orbit-induced spin splitting in two-dimensional transition metal dichalcogenides semiconductors | The established spin splitting in monolayer (ML) of transition metal
dichalcogenides (TMDs) that is caused by inversion symmetry breaking is
dictated by mirror symmetry operations to exhibit fully out-of-plane direction
of spin polarization. Through first-principles density functional theory
calculations, we show that polarity-induced mirror symmetry breaking leads to
new sizable spin splitting having in-plane spin polarization. These splittings
are effectively controlled by tuning the polarity using biaxial strain.
Furthermore, the admixtures of the out-of-plane and in-plane spin-polarized
states in the strained polar systems are identified, which is expected to
influence the spin relaxation through the Dyakonov-Perel mechanism. Our study
clarified that the polarity-induced mirror symmetry breaking plays an important
role in controlling the spin splitting and spin relaxation in the TMDs ML,
which is useful for designing future spintronic devices.
| 0 | 1 | 0 | 0 | 0 | 0 |
Some parametrized dynamic priority policies for 2-class M/G/1 queues: completeness and applications | Completeness of a dynamic priority scheduling scheme is of fundamental
importance for the optimal control of queues in areas as diverse as computer
communications, communication networks, supply chains and manufacturing
systems. Our first main contribution is to identify the mean waiting time
completeness as a unifying aspect for four different dynamic priority
scheduling schemes by proving their completeness and equivalence in 2-class
M/G/1 queue. These dynamic priority schemes are earliest due date based, head
of line priority jump, relative priority, and probabilistic priority.
In our second main contribution, we characterize the optimal scheduling
policies for the case studies in different domains by exploiting the
completeness of above dynamic priority schemes. The major theme of second main
contribution is resource allocation/optimal control in revenue management
problems for contemporary systems such as cloud computing, high-performance
computing, etc., where congestion is inherent. Using completeness and
theoretically tractable nature of relative priority policy, we study the impact
of approximation in a fairly generic data network utility framework. We
introduce the notion of min-max fairness in multi-class queues and show that a
simple global FCFS policy is min-max fair. Next, we re-derive the celebrated
$c/\rho$ rule for 2-class M/G/1 queues by an elegant argument and also simplify
a complex joint pricing and scheduling problem for a wider class of scheduling
policies.
| 1 | 0 | 0 | 0 | 0 | 0 |
Vector bundles over classifying spaces of p-local finite groups and Benson-Carlson duality | In this paper we obtain a description of the Grothendieck group of complex
vector bundles over the classifying space of a p-local finite group in terms of
representation rings of subgroups of its Sylow. We also prove a stable elements
formula for generalized cohomological invariants of p-local finite groups,
which is used to show the existence of unitary embeddings of p-local finite
groups. Finally, we show that the augmentation map for the cochains of the
classifying space of a p-local finite group is Gorenstein in the sense of
Dwyer-Greenlees-Iyengar and obtain some consequences about the cohomology ring
of these classifying spaces.
| 0 | 0 | 1 | 0 | 0 | 0 |
Identification of a complete YPT1 Rab GTPase sequence from the fungal pathogen Colletotrichum incanum | Colletotrichum represent a genus of fungal species primarily known as plant
pathogens with severe economic impacts in temperate, subtropical and tropical
climates Consensus taxonomy and classification systems for Colletotrichum
species have been undergoing revision as high resolution genomic data becomes
available. Here we propose an alternative annotation that provides a complete
sequence for a Colletotrichum YPT1 gene homolog using the whole genome shotgun
sequence of Colletotrichum incanum isolated from soybean crops in Illinois,
USA.
| 0 | 0 | 0 | 0 | 1 | 0 |
Multiplicative Structure in the Stable Splitting of $ΩSL_n(\mathbb{C})$ | The space of based loops in $SL_n(\mathbb{C})$, also known as the affine
Grassmannian of $SL_n(\mathbb{C})$, admits an $\mathbb{E}_2$ or fusion product.
Work of Mitchell and Richter proves that this based loop space stably splits as
an infinite wedge sum. We prove that the Mitchell--Richter splitting is
coherently multiplicative, but not $\mathbb{E}_2$. Nonetheless, we show that
the splitting becomes $\mathbb{E}_2$ after base-change to complex cobordism.
Our proof of the $\mathbb{A}_\infty$ splitting involves on the one hand an
analysis of the multiplicative properties of Weiss calculus, and on the other a
use of Beilinson--Drinfeld Grassmannians to verify a conjecture of Mahowald and
Richter. Other results are obtained by explicit, obstruction-theoretic
computations.
| 0 | 0 | 1 | 0 | 0 | 0 |
Strange duality on rational surfaces II: higher rank cases | We study Le Potier's strange duality conjecture on a rational surface. We
focus on the strange duality map $SD_{c_n^r,L}$ which involves the moduli space
of rank $r$ sheaves with trivial first Chern class and second Chern class $n$,
and the moduli space of 1-dimensional sheaves with determinant $L$ and Euler
characteristic 0. We show there is an exact sequence relating the map
$SD_{c_r^r,L}$ to $SD_{c^{r-1}_{r},L}$ and $SD_{c_r^r,L\otimes K_X}$ for all
$r\geq1$ under some conditions on $X$ and $L$ which applies to a large number
of cases on $\p^2$ or Hirzebruch surfaces . Also on $\mathbb{P}^2$ we show that
for any $r>0$, $SD_{c^r_r,dH}$ is an isomorphism for $d=1,2$, injective for
$d=3$ and moreover $SD_{c_3^3,rH}$ and $SD_{c_3^2,rH}$ are injective. At the
end we prove that the map $SD_{c_n^2,L}$ ($n\geq2$) is an isomorphism for
$X=\mathbb{P}^2$ or Fano rational ruled surfaces and $g_L=3$, and hence so is
$SD_{c_3^3,L}$ as a corollary of our main result.
| 0 | 0 | 1 | 0 | 0 | 0 |
Graph learning under sparsity priors | Graph signals offer a very generic and natural representation for data that
lives on networks or irregular structures. The actual data structure is however
often unknown a priori but can sometimes be estimated from the knowledge of the
application domain. If this is not possible, the data structure has to be
inferred from the mere signal observations. This is exactly the problem that we
address in this paper, under the assumption that the graph signals can be
represented as a sparse linear combination of a few atoms of a structured graph
dictionary. The dictionary is constructed on polynomials of the graph
Laplacian, which can sparsely represent a general class of graph signals
composed of localized patterns on the graph. We formulate a graph learning
problem, whose solution provides an ideal fit between the signal observations
and the sparse graph signal model. As the problem is non-convex, we propose to
solve it by alternating between a signal sparse coding and a graph update step.
We provide experimental results that outline the good graph recovery
performance of our method, which generally compares favourably to other recent
network inference algorithms.
| 1 | 0 | 0 | 1 | 0 | 0 |
Improving Trajectory Optimization using a Roadmap Framework | We present an evaluation of several representative sampling-based and
optimization-based motion planners, and then introduce an integrated motion
planning system which incorporates recent advances in trajectory optimization
into a sparse roadmap framework. Through experiments in 4 common application
scenarios with 5000 test cases each, we show that optimization-based or
sampling-based planners alone are not effective for realistic problems where
fast planning times are required. To the best of our knowledge, this is the
first work that presents such a systematic and comprehensive evaluation of
state-of-the-art motion planners, which are based on a significant amount of
experiments. We then combine different stand-alone planners with trajectory
optimization. The results show that the combination of our sparse roadmap and
trajectory optimization provides superior performance over other standard
sampling-based planners combinations. By using a multi-query roadmap instead of
generating completely new trajectories for each planning problem, our approach
allows for extensions such as persistent control policy information associated
with a trajectory across planning problems. Also, the sub-optimality resulting
from the sparsity of roadmap, as well as the unexpected disturbances from the
environment, can both be overcome by the real-time trajectory optimization
process.
| 1 | 0 | 0 | 0 | 0 | 0 |
Smooth and Efficient Policy Exploration for Robot Trajectory Learning | Many policy search algorithms have been proposed for robot learning and
proved to be practical in real robot applications. However, there are still
hyperparameters in the algorithms, such as the exploration rate, which requires
manual tuning. The existing methods to design the exploration rate manually or
automatically may not be general enough or hard to apply in the real robot. In
this paper, we propose a learning model to update the exploration rate
adaptively. The overall algorithm is a combination of methods proposed by other
researchers. Smooth trajectories for the robot can be produced by the algorithm
and the updated exploration rate maximizes the lower bound of the expected
return. Our method is tested in the ball-in-cup problem. The results show that
our method can receive the same learning outcome as the previous methods but
with fewer iterations.
| 1 | 0 | 0 | 0 | 0 | 0 |
Social Innovation and the Evolution of Creative, Sustainable Worldviews | The ideas that we forge creatively as individuals and groups build on one
another in a manner that is cumulative and adaptive, forming open-ended
lineages across space and time. Thus, human culture is believed to evolve. The
pervasiveness of cross-domain creativity--as when a song inspires a
painting--would appear indicative of discontinuities in cultural lineages.
However, if what evolves through culture is our worldviews--the webs of
thoughts, ideas, and attitudes that constitutes our way of seeing being in the
world--then the problem of discontinuities is solved. The state of a worldview
can be affected by information assimilated in one domain, and this
change-of-state can be expressed in another domain. In this view, the gesture,
narrative, or artifact that constitutes a specific creative act is not what is
evolving; it is merely the external manifestation of the state of an evolving
worldview. Like any evolutionary process, cultural evolution requires a balance
between novelty, via the generation of variation, and continuity, via the
preservation of variants that are adaptive. In cultural evolution, novelty is
generated through creativity, and continuity is provided by social learning
processes, e.g., imitation. Both the generative and imitative aspects of
cultural evolution are affected by social media. We discuss the trajectory from
social ideation to social innovation, focusing on the role of
self-organization, renewal, and perspective-taking at the individual and social
group level.
| 0 | 0 | 0 | 0 | 1 | 0 |
Interaction between magnetic moments and itinerant carriers in d0 ferromagnetic SiC | Elucidating the interaction between magnetic moments and itinerant carriers
is an important step to spintronic applications. Here, we investigate magnetic
and transport properties in d0 ferromagnetic SiC single crystals prepared by
postimplantation pulsed laser annealing. Magnetic moments are contributed by
the p states of carbon atoms, but their magnetic circular dichroism is
different from that in semi-insulating SiC samples. The anomalous Hall effect
and negative magnetoresistance indicate the influence of d0 spin order on free
carriers. The ferromagnetism is relatively weak in N-implanted SiC compared
with that in Al-implanted SiC after annealing. The results suggest that d0
magnetic moments and itinerant carriers can interact with each other, which
will facilitate the development of SiC spintronic devices with d0
ferromagnetism.
| 0 | 1 | 0 | 0 | 0 | 0 |
Il Fattore di Sylvester | Sylvester factor, an essential part of the asymptotic formula of Hardy and
Littlewood which is the extended Goldbach conjecture, regarded as strongly
multiplicative arithmetic function, has several remarkable properties.
| 0 | 0 | 1 | 0 | 0 | 0 |
Bright and Gap Solitons in Membrane-Type Acoustic Metamaterials | We study analytically and numerically envelope solitons (bright and gap
solitons) in a one-dimensional, nonlinear acoustic metamaterial, composed of an
air-filled waveguide periodically loaded by clamped elastic plates. Based on
the transmission line approach, we derive a nonlinear dynamical lattice model
which, in the continuum approximation, leads to a nonlinear, dispersive and
dissipative wave equation. Applying the multiple scales perturbation method, we
derive an effective lossy nonlinear Schrödinger equation and obtain
analytical expressions for bright and gap solitons. We also perform direct
numerical simulations to study the dissipation-induced dynamics of the bright
and gap solitons. Numerical and analytical results, relying on the analytical
approximations and perturbation theory for solions, are found to be in good
agreement.
| 0 | 1 | 0 | 0 | 0 | 0 |
Local reservoir model for choice-based learning | Decision making based on behavioral and neural observations of living systems
has been extensively studied in brain science, psychology, and other
disciplines. Decision-making mechanisms have also been experimentally
implemented in physical processes, such as single photons and chaotic lasers.
The findings of these experiments suggest that there is a certain common basis
in describing decision making, regardless of its physical realizations. In this
study, we propose a local reservoir model to account for choice-based learning
(CBL). CBL describes decision consistency as a phenomenon where making a
certain decision increases the possibility of making that same decision again
later, which has been intensively investigated in neuroscience, psychology,
etc. Our proposed model is inspired by the viewpoint that a decision is
affected by its local environment, which is referred to as a local reservoir.
If the size of the local reservoir is large enough, consecutive decision making
will not be affected by previous decisions, thus showing lower degrees of
decision consistency in CBL. In contrast, if the size of the local reservoir
decreases, a biased distribution occurs within it, which leads to higher
degrees of decision consistency in CBL. In this study, an analytical approach
on local reservoirs is presented, as well as several numerical demonstrations.
Furthermore, a physical architecture for CBL based on single photons is
discussed, and the effects of local reservoirs is numerically demonstrated.
Decision consistency in human decision-making tasks and in recruiting empirical
data are evaluated based on local reservoir. In summary, the proposed local
reservoir model paves a path toward establishing a foundation for computational
mechanisms and the systematic analysis of decision making on different levels.
| 0 | 0 | 0 | 1 | 1 | 0 |
Weil-Petersson geometry on the space of Bridgeland stability conditions | Inspired by mirror symmetry, we investigate some differential geometric
aspects of the space of Bridgeland stability conditions on a Calabi-Yau
triangulated category. The aim is to develop theory of Weil-Petersson geometry
on the stringy Kähler moduli space. A few basic examples are studied. In
particular, we identify our Weil-Petersson metric with the Bergman metric on a
Siegel modular variety in the case of the self-product of an elliptic curve.
| 0 | 0 | 1 | 0 | 0 | 0 |
Analysis of a Sputtered Si Surface for Ar Sputter Gas Supply Purity Monitoring | For sputter depth profiling often sample erosion by Ar+ ions is used. Only a
high purity of the sputter gas and a low contamination level of the ion gun
avoids misleading depth profile measurements results. Here a new measurement
procedure is presented, which monitors these parameters. A Si sample is
sputtered inside the instrument and then the surface concentration of the
elements Ar, C, N and O is measured. Results of such measurements of an XPS
microprobe PHI Quantum 2000, which were recorded over a period of 10 years, are
presented.
| 0 | 1 | 0 | 0 | 0 | 0 |
Extracting spectroscopic molecular parameters from short pulse photo-electron angular distributions | Using a quantum wave packet simulation including the nuclear and electronic
degrees of freedom, we investigate the femtosecond and picosecond energy- and
angle-resolved photoelectron spectra of the E($^1\Sigma_g^+$) electronic state
of Li$_2$. We find that the angular distributions of the emitted photoelectrons
depend strongly on the pulse duration in the regime of ultrashort laser pulses.
This effect is illustrated by the extraction of a time-dependent asymmetry
parameter whose variation with pulse duration can be explained by an incoherent
average over different ion rotational quantum numbers. We then derive for the
variation of the asymmetry parameter a simple analytical formula, which can be
used to extract the asymptotic CW asymmetry parameters of individual
transitions from measurements performed with ultra-short pulses.
| 0 | 1 | 0 | 0 | 0 | 0 |
A Generative Model for Score Normalization in Speaker Recognition | We propose a theoretical framework for thinking about score normalization,
which confirms that normalization is not needed under (admittedly fragile)
ideal conditions. If, however, these conditions are not met, e.g. under
data-set shift between training and runtime, our theory reveals dependencies
between scores that could be exploited by strategies such as score
normalization. Indeed, it has been demonstrated over and over experimentally,
that various ad-hoc score normalization recipes do work. We present a first
attempt at using probability theory to design a generative score-space
normalization model which gives similar improvements to ZT-norm on the
text-dependent RSR 2015 database.
| 1 | 0 | 0 | 1 | 0 | 0 |
Deep Networks tag the location of bird vocalisations on audio spectrograms | This work focuses on reliable detection and segmentation of bird
vocalizations as recorded in the open field. Acoustic detection of avian sounds
can be used for the automatized monitoring of multiple bird taxa and querying
in long-term recordings for species of interest. These tasks are tackled in
this work, by suggesting two approaches: A) First, DenseNets are applied to
weekly labeled data to infer the attention map of the dataset (i.e. Salience
and CAM). We push further this idea by directing attention maps to the YOLO v2
Deepnet-based, detection framework to localize bird vocalizations. B) A deep
autoencoder, namely the U-net, maps the audio spectrogram of bird vocalizations
to its corresponding binary mask that encircles the spectral blobs of
vocalizations while suppressing other audio sources. We focus solely on
procedures requiring minimum human attendance, suitable to scan massive volumes
of data, in order to analyze them, evaluate insights and hypotheses and
identify patterns of bird activity. Hopefully, this approach will be valuable
to researchers, conservation practitioners, and decision makers that need to
design policies on biodiversity issues.
| 1 | 0 | 0 | 0 | 0 | 0 |
To Wait or Not to Wait: Two-way Functional Hazards Model for Understanding Waiting in Call Centers | Telephone call centers offer a convenient communication channel between
businesses and their customers. Efficient management of call centers needs
accurate modeling of customer waiting behavior, which contains important
information about customer patience (how long a customer is willing to wait)
and service quality (how long a customer needs to wait to get served). Hazard
functions offer dynamic characterization of customer waiting behavior, and
provide critical inputs for agent scheduling. Motivated by this application, we
develop a two-way functional hazards (tF-Hazards) model to study customer
waiting behavior as a function of two timescales, waiting duration and the time
of day that a customer calls in. The model stems from a two-way piecewise
constant hazard function, and imposes low-rank structure and smoothness on the
hazard rates to enhance interpretability. We exploit an alternating direction
method of multipliers (ADMM) algorithm to optimize a penalized likelihood
function of the model. We carefully analyze the data from a US bank call
center, and provide informative insights about customer patience and service
quality patterns along waiting time and across different times of a day. The
findings provide primitive inputs for call center agent staffing and
scheduling, as well as for call center practitioners to understand the effect
of system protocols on customer waiting behavior.
| 0 | 0 | 0 | 1 | 0 | 0 |
Bayesian Gaussian models for interpolating large-dimensional data at misaligned areal units | Areal level spatial data are often large, sparse and may appear with
geographical shapes that are regular or irregular (e.g., postcode). Moreover,
sometimes it is important to obtain predictive inference in regular or
irregular areal shapes that is misaligned with the observed spatial areal
geographical boundary. For example, in a survey the respondents were asked
about their postcode, however for policy making purposes, researchers are often
interested to obtain information at the SA2. The statistical challenge is to
obtain spatial prediction at the SA2s, where the SA2s may have overlapped
geographical boundaries with postcodes. The study is motivated by a practical
survey data obtained from the Australian National University (ANU) Poll. Here
the main research question is to understand respondents' satisfaction level
with the way Australia is heading. The data are observed at 1,944 postcodes
among the 2,516 available postcodes across Australia, and prediction is
obtained at the 2,196 SA2s. The proposed method also explored through a
grid-based simulation study, where data have been observed in a regular grid
and spatial prediction has been done in a regular grid that has a misaligned
geographical boundary with the first regular grid-set. The real-life example
with ANU Poll data addresses the situation of irregular geographical boundaries
that are misaligned, i.e., model fitted with postcode data and hence obtained
prediction at the SA2. A comparison study is also performed to validate the
proposed method. In this paper, a Gaussian model is constructed under Bayesian
hierarchy. The novelty lies in the development of the basis function that can
address spatial sparsity and localised spatial structure. It can also address
the large-dimensional spatial data modelling problem by constructing knot based
reduced-dimensional basis functions.
| 0 | 0 | 0 | 1 | 0 | 0 |
Report on TBAS 2012: Workshop on Task-Based and Aggregated Search | The ECIR half-day workshop on Task-Based and Aggregated Search (TBAS) was
held in Barcelona, Spain on 1 April 2012. The program included a keynote talk
by Professor Jarvelin, six full paper presentations, two poster presentations,
and an interactive discussion among the approximately 25 participants. This
report overviews the aims and contents of the workshop and outlines the major
outcomes.
| 1 | 0 | 0 | 0 | 0 | 0 |
Reliability of the measured velocity anisotropy of the Milky Way stellar halo | Determining the velocity distribution of halo stars is essential for
estimating the mass of the Milky Way and for inferring its formation history.
Since the stellar halo is a dynamically hot system, the velocity distribution
of halo stars is well described by the 3-dimensional velocity dispersions
$(\sigma_r, \sigma_\theta, \sigma_\phi)$, or by the velocity anisotropy
parameter $\beta=1-(\sigma_\theta^2+\sigma_\phi^2)/(2\sigma_r^2)$. Direct
measurements of $(\sigma_r, \sigma_\theta, \sigma_\phi)$ consistently suggest
$\beta =0.5$-$0.7$ for nearby halo stars. In contrast, the value of $\beta$ at
large Galactocentric radius $r$ is still controversial, since reliable proper
motion data are available for only a handful of stars. In the last decade,
several authors have tried to estimate $\beta$ for distant halo stars by
fitting the observed line-of-sight velocities at each radius with simple
velocity distribution models (local fitting methods). Some results of local
fitting methods imply $\beta<0$ at $r \gtrsim 20 \;\rm{kpc}$, which is
inconsistent with recent predictions from cosmological simulations. Here we
perform mock-catalogue analyses to show that the estimates of $\beta$ based on
local fitting methods are reliable only at $r \leq 15 \;\rm{kpc}$ with the
current sample size ($\sim10^3$ stars at a given radius). As $r$ increases, the
line-of-sight velocity (corrected for the Solar reflex motion) becomes
increasingly closer to the Galactocentric radial velocity, so that it becomes
increasingly more difficult to estimate tangential velocity dispersion
$(\sigma_\theta, \sigma_\phi)$ from line-of-sight velocity distribution. Our
results suggest that the forthcoming Gaia data will be crucial for
understanding the velocity distribution of halo stars at $r \geq 20\;\rm{kpc}$.
| 0 | 1 | 0 | 0 | 0 | 0 |
$H$-compactness of elliptic operators on weighted Riemannian Manifolds | In this paper we study the asymptotic behavior of second-order uniformly
elliptic operators on weighted Riemannian manifolds. We appeal to the notion of
\mbox{$H$-convergence} introduced by Murat and Tartar. In our main result we
establish an \mbox{$H$-compactness} result that applies to elliptic operators
with measurable, uniformly elliptic coefficients on weighted Riemannian
manifolds. We further discuss the special case of "locally periodic"
coefficients and study the asymptotic behavior of the Laplace-Beltrami operator
on families of weighted manifolds obtained from a reference manifold by a
conformal (rapidly oscillating) change of metrics.
| 0 | 0 | 1 | 0 | 0 | 0 |
$b$-symbol distance distribution of repeated-root cyclic codes | Symbol-pair codes, introduced by Cassuto and Blaum [1], have been raised for
symbol-pair read channels. This new idea is motivated by the limitation of the
reading process in high-density data storage technologies. Yaakobi et al. [8]
introduced codes for $b$-symbol read channels, where the read operation is
performed as a consecutive sequence of $b>2$ symbols. In this paper, we come up
with a method to compute the $b$-symbol-pair distance of two $n$-tuples, where
$n$ is a positive integer. Also, we deal with the $b$-symbol-pair distances of
some kind of cyclic codes of length $p^e$ over $\mathbb{F}_{p^m}$.
| 1 | 0 | 0 | 0 | 0 | 0 |
Knowledge Fusion via Embeddings from Text, Knowledge Graphs, and Images | We present a baseline approach for cross-modal knowledge fusion. Different
basic fusion methods are evaluated on existing embedding approaches to show the
potential of joining knowledge about certain concepts across modalities in a
fused concept representation.
| 1 | 0 | 0 | 1 | 0 | 0 |
Observation of spin superfluidity: YIG magnetic films and beyond | From topology of the order parameter of the magnon condensate observed in
yttrium-iron-garnet (YIG) magnetic films one must not expect energetic barriers
making spin supercurrents metastable. But we show that some barriers of
dynamical origin are possible nevertheless until the gradient of the phase
(angle of spin precession) does not exceed the critical value (analog of the
Landau critical velocity in superfluids). On the other hand, recently published
claims of experimental detection of spin superfluidity in YIG films and
antiferromagnets are not justified, and spin superfluidity in magnetically
ordered solids has not yet been experimentally confirmed.
| 0 | 1 | 0 | 0 | 0 | 0 |
Tuning of Interlayer Coupling in Large-Area Graphene/WSe2 van der Waals Heterostructure via Ion Irradiation: Optical Evidences and Photonic Applications | Van der Waals (vdW) heterostructures are receiving great attentions due to
their intriguing properties and potentials in many research fields. The flow of
charge carriers in vdW heterostructures can be efficiently rectified by the
inter-layer coupling between neighboring layers, offering a rich collection of
functionalities and a mechanism for designing atomically thin devices.
Nevertheless, non-uniform contact in larger-area heterostructures reduces the
device efficiency. In this work, ion irradiation had been verified as an
efficient technique to enhance the contact and interlayer coupling in the newly
developed graphene/WSe2 hetero-structure with a large area of 10 mm x 10 mm.
During the ion irradiation process, the morphology of monolayer graphene had
been modified, promoting the contact with WSe2. Experimental evidences of the
tunable interlayer electron transfer are displayed by investigation of
photoluminescence and ultrafast absorption of the irradiated heterostructure.
Besides, we have found that in graphene/WSe2 heterostructure, graphene serves
as a fast channel for the photo-excited carriers to relax in WSe2, and the
nonlinear absorption of WSe2 could be effectively tuned by the carrier transfer
process in graphene, enabling specific optical absorption of the
heterostructure in comparison with separated graphene or WSe2. On the basis of
these new findings, by applying the ion beam modified graphene/WSe2
heterostructure as a saturable absorber, Q-switched pulsed lasing with
optimized performance has been realized in a Nd:YAG waveguide cavity. This work
paves the way towards developing novel devices based on large-area
heterostructures by using ion beam irradiation.
| 0 | 1 | 0 | 0 | 0 | 0 |
Learning to Multi-Task by Active Sampling | One of the long-standing challenges in Artificial Intelligence for learning
goal-directed behavior is to build a single agent which can solve multiple
tasks. Recent progress in multi-task learning for goal-directed sequential
problems has been in the form of distillation based learning wherein a student
network learns from multiple task-specific expert networks by mimicking the
task-specific policies of the expert networks. While such approaches offer a
promising solution to the multi-task learning problem, they require supervision
from large expert networks which require extensive data and computation time
for training. In this work, we propose an efficient multi-task learning
framework which solves multiple goal-directed tasks in an on-line setup without
the need for expert supervision. Our work uses active learning principles to
achieve multi-task learning by sampling the harder tasks more than the easier
ones. We propose three distinct models under our active sampling framework. An
adaptive method with extremely competitive multi-tasking performance. A
UCB-based meta-learner which casts the problem of picking the next task to
train on as a multi-armed bandit problem. A meta-learning method that casts the
next-task picking problem as a full Reinforcement Learning problem and uses
actor critic methods for optimizing the multi-tasking performance directly. We
demonstrate results in the Atari 2600 domain on seven multi-tasking instances:
three 6-task instances, one 8-task instance, two 12-task instances and one
21-task instance.
| 1 | 0 | 0 | 0 | 0 | 0 |
Sufficient Conditions for Idealised Models to Have No Adversarial Examples: a Theoretical and Empirical Study with Bayesian Neural Networks | We prove, under two sufficient conditions, that idealised models can have no
adversarial examples. We discuss which idealised models satisfy our conditions,
and show that idealised Bayesian neural networks (BNNs) satisfy these. We
continue by studying near-idealised BNNs using HMC inference, demonstrating the
theoretical ideas in practice. We experiment with HMC on synthetic data derived
from MNIST for which we know the ground-truth image density, showing that
near-perfect epistemic uncertainty correlates to density under image manifold,
and that adversarial images lie off the manifold in our setting. This suggests
why MC dropout, which can be seen as performing approximate inference, has been
observed to be an effective defence against adversarial examples in practice;
We highlight failure-cases of non-idealised BNNs relying on dropout, suggesting
a new attack for dropout models and a new defence as well. Lastly, we
demonstrate the defence on a cats-vs-dogs image classification task with a
VGG13 variant.
| 0 | 0 | 0 | 1 | 0 | 0 |
Long term availability of raw experimental data in experimental fracture mechanics | Experimental data availability is a cornerstone for reproducibility in
experimental fracture mechanics, which is crucial to the scientific method.
This short communication focuses on the accessibility and long term
availability of raw experimental data. The corresponding authors of the eleven
most cited papers, related to experimental fracture mechanics, for every year
from 2000 up to 2016, were kindly asked about the availability of the raw
experimental data associated with each publication. For the 187 e-mails sent:
22.46% resulted in outdated contact information, 57.75% of the authors did
received our request and did not reply, and 19.79 replied to our request. The
availability of data is generally low with only $11$ available data sets
(5.9%). The authors identified two main issues for the lacking availability of
raw experimental data. First, the ability to retrieve data is strongly attached
to the the possibility to contact the corresponding author. This study suggests
that institutional e-mail addresses are insufficient means for obtaining
experimental data sets. Second, lack of experimental data is also due that
submission and publication does not require to make the raw experimental data
available. The following solutions are proposed: (1) Requirement of unique
identifiers, like ORCID or ResearcherID, to detach the author(s) from their
institutional e-mail address, (2) Provide DOIs, like Zenodo or Dataverse, to
make raw experimental data citable, and (3) grant providing organizations
should ensure that experimental data by public funded projects is available to
the public.
| 0 | 0 | 0 | 1 | 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.