October 2020 arXiv papers — page 59

Xiaozheng Chen, Xueliang Li

Let $G$ be a graph of order $n$ with an edge-coloring $c$, and let $δ^c(G)$ denote the minimum color degree of $G$. A subgraph $F$ of $G$ is called rainbow if all edges of $F$ have pairwise distinct colors. There have been a lot results on rainbow cycles of edge-colored graphs. In this paper, we show that (i) if $δ^c(G)>\frac{3n-3}{4}$, then every vertex of

Vera Mikyoung Hur, Zhao Yang

We investigate the spectral instability of a $2\pi/\kappa$ periodic Stokes wave of sufficiently small amplitude, traveling in water of unit depth, under gravity. Numerical evidence suggests instability whenever the unperturbed wave is resonant with its infinitesimal perturbations. This has not been analytically studied except for the Benjamin--Feir instabili

Ha Thi Phuong Thao, Balamurali B. T., Dorien Herremans, Gemma Roig

In this work, we propose different variants of the self-attention based network for emotion prediction from movies, which we call AttendAffectNet. We take both audio and video into account and incorporate the relation among multiple modalities by applying self-attention mechanism in a novel manner into the extracted features for emotion prediction. We compar

Chen Yang, Zhigang Song, Xiaotian Sun, Jing Lu

For a long time, two-dimensional (2D) hexagonal MoS2 was proposed as a promising material for valleytronic system. However, the limited size of growth and low carrier motilities in MoS2 restrict its further application. Very recently, a new kind of hexagonal 2D MXene, MoSi2N4, was successfully synthesized with large size, excellent ambient stability, and con

Joseph N Stember, Hrithwik Shalu

Purpose Supervised deep learning in radiology suffers from notorious inherent limitations: 1) It requires large, hand-annotated data sets, 2) It is non-generalizable, and 3) It lacks explainability and intuition. We have recently proposed Reinforcement Learning to address all threes. However, we applied it to images with radiologist eye tracking points, whic

Somenath Pal, Guruprasad Kadam, Hiranmaya Mishra, Abhijit Bhattacharyya

We investigate the effects of repulsive interaction between hadrons on the fluctuations of the conserved charges. We calculate the baryon,the electric charge and the strangeness susceptibilities within the ambit of hadron resonance gas model extended to include the short range repulsive interactions.The repulsive interactions are included through a mean-fiel

Yangyang Shi, Yongqiang Wang, Chunyang Wu, Ching-Feng Yeh

This paper proposes an efficient memory transformer Emformer for low latency streaming speech recognition. In Emformer, the long-range history context is distilled into an augmented memory bank to reduce self-attention's computation complexity. A cache mechanism saves the computation for the key and value in self-attention for the left context. Emformer appl

Yan Zhang, Jin-Hui Wu, M. Artoni, G. C. La Rocca

We propose a model for realizing frequency-dependent spatial variations of the probe susceptibility in a cold atomic sample. It is found that the usual Kramers-Kronig (KK) relation between real and imaginary parts of the probe susceptibility in the frequency domain can be mapped into the space domain as a far detuned control field of intensity linearly varie

Venkata Pavan Kumar Miriyala, Masatoshi Ishii

Embedding artificial intelligence at the edge (edge-AI) is an elegant solution to tackle the power and latency issues in the rapidly expanding Internet of Things. As edge devices typically spend most of their time in sleep mode and only wake-up infrequently to collect and process sensor data, non-volatile in-memory computing (NVIMC) is a promising approach t

Kwan-Lok Li, Franz-Josef Hambsch, Ulisse Munari, Brian D. Metzger

We report on the Fermi-LAT detection (with ~5.7 sigma significance) as well as the multi-wavelength analysis of the 2017 nova eruption V549 Vel. Unlike the recent shock-powered novae ASASSN-16ma and V906 Car, the optical and gamma-ray light curves of V549 Vel show no correlation, likely implying relatively weak shocks in the eruption. Gaia detected a candida

Bohan Chen, Chang-Han Rhee, Bert Zwart

For a class of additive processes driven by the affine recursion $X_{n+1} = A_n X_n + B_n$, we develop a sample-path large deviations principle in the $M_1'$ topology on $D [0,1]$. We allow $B_n$ to have both signs and focus on the case where Kesten's condition holds on $A_1$, leading to heavy-tailed distributions. The most likely paths in our large deviatio

Xiaoguang Luo, Hexin Zhang, Dan Liu, Nannan Han

Efficiency at maximum power (EMP) is a very important specification for a heat engine to evaluate the capacity of outputting adequate power with high efficiency. It has been proved theoretically that the limit EMP of thermoelectric heat engine can be achieved with the hypothetical boxcar-shaped electron transmission, which is realized here by the resonant tu

Kang Li, Ján Špakula, Jiawen Zhang

Our main result about rigidity of Roe algebras is the following: if $X$ and $Y$ are metric spaces with bounded geometry such that their Roe algebras are $*$-isomorphic, then $X$ and $Y$ are coarsely equivalent provided that either $X$ or $Y$ contains no sparse subspaces consisting of ghostly measured asymptotic expanders. Note that this geometric condition g

J. Heinzel, M. W. Coughlin, T. Dietrich, M. Bulla

The detection of AT2017gfo proved that binary neutron star mergers are progenitors of kilonovae. Using a combination of numerical-relativity and radiative-transfer simulations, the community has developed sophisticated models for these transients for a wide portion of the expected parameter space. Using these simulations and surrogate models made from them,

Alex Cowan

We describe an algorithm that we used to compute the q-expansions of all weight 2 cusp forms of prime level at most 2,000,000 and dimension at most 6. We also present an algorithm that we used to verify that there was only one cusp form of dimension 7 or more per Atkin-Lehner eigenspace for prime levels between 10,000 and 1,000,000. Our algorithm is based on

Jianhao Yan, Fandong Meng, Jie Zhou

Transformer models achieve remarkable success in Neural Machine Translation. Many efforts have been devoted to deepening the Transformer by stacking several units (i.e., a combination of Multihead Attentions and FFN) in a cascade, while the investigation over multiple parallel units draws little attention. In this paper, we propose the Multi-Unit Transformer

Mahdi Abolghasemi, Rob J Hyndman, Evangelos Spiliotis, Christoph Bergmeir

Model selection has been proven an effective strategy for improving accuracy in time series forecasting applications. However, when dealing with hierarchical time series, apart from selecting the most appropriate forecasting model, forecasters have also to select a suitable method for reconciling the base forecasts produced for each series to make sure they

Erica Hammerstein, Suvi Gezari, Sjoert van Velzen, S. Bradley Cenko

We study the properties of the galaxies hosting the first 19 tidal disruption events (TDEs) detected with the Zwicky Transient Facility (ZTF) within the context of a carefully constructed, representative host galaxy sample. We find that the ZTF sample of TDE hosts is dominated by compact "green valley" galaxies. After we restrict the comparison sample to gal

Norman J. Morgenstern Horing

This work addresses the statistical thermodynamic features of a Landau-quantized T-3 Dice lattice in a normal magnetic field in the nondegenerate statistical regime. Our study includes analyses of the Grand Potential, magnetic moment, entropy and specific heat at constant volume, with explicit determination of their temperature and magnetic field dependencie

Mladen Bestvina, Ryan Dickmann, George Domat, Sanghoon Kwak

The far-reaching work of Dahmani-Guirardel-Osin and recent work of Clay-Mangahas-Margalit provide geometric approaches to the study of the normal closure of a subgroup (or a collection of subgroups)in an ambient group $G$. Their work gives conditions under which the normal closure in $G$ is a free product. In this paper we unify their results and simplify an

Yehui Tang, Yunhe Wang, Yixing Xu, Dacheng Tao

This paper proposes a reliable neural network pruning algorithm by setting up a scientific control. Existing pruning methods have developed various hypotheses to approximate the importance of filters to the network and then execute filter pruning accordingly. To increase the reliability of the results, we prefer to have a more rigorous research design by inc

Ying-Hua Yue, Sheng-Li Qin, Tie Liu, Meng-Yao Tang

Thirty massive clumps associated with bright infrared sources were observed to detect the infall signatures and characterize infall properties in the envelope of the massive clumps by APEX telescope in CO(4-3) and C$^{17}$O(3-2) lines. Eighteen objects have "blue profile" in CO(4-3) line with virial parameters less than 2, suggesting that global collapse is

Daomin Cao, Qing Guo, Changjun Zou

In this paper, we study the Cauchy problem of the nonlinear Schrödinger equation with a nontrival potential $V_\varepsilon(x)$. In particular, we consider the case where the initial data is close to a superposition of $k$ solitons with prescribed phase and location, and investigate the evolution of the Schrödinger system. We prove that over a large time inte

Xiangguo Sun, Hongzhi Yin, Bo Liu, Hongxu Chen

Recently, graph neural networks have been widely used for network embedding because of their prominent performance in pairwise relationship learning. In the real world, a more natural and common situation is the coexistence of pairwise relationships and complex non-pairwise relationships, which is, however, rarely studied. In light of this, we propose a grap

Jennifer Williams, Yi Zhao, Erica Cooper, Junichi Yamagishi

We present a new approach to disentangle speaker voice and phone content by introducing new components to the VQ-VAE architecture for speech synthesis. The original VQ-VAE does not generalize well to unseen speakers or content. To alleviate this problem, we have incorporated a speaker encoder and speaker VQ codebook that learns global speaker characteristics

Jesus Tordesillas, Jonathan P. How

This paper studies the polynomial basis that generates the smallest $n$-simplex enclosing a given $n^{\text{th}}$-degree polynomial curve in $\mathbb{R}^n$. Although the Bernstein and B-Spline polynomial bases provide feasible solutions to this problem, the simplexes obtained by these bases are not the smallest possible, which leads to overly conservative re

Xiaoling Cui, Yinfeng Ma

We study the stability of quantum droplet and its associated phase transitions in ultracold Bose-Bose mixtures uniformly confined in quasi-two-dimension. We show that the confinement-induced boundary effect can be significant when increasing the atom number or reducing the confinement length, which destabilizes the quantum droplet towards the formation of a

Yong Hu, Tong Zhang

We prove the optimal Noether-Severi inequality that $\mathrm{vol}(X) \ge \frac{4}{3} \chi(\omega_{X})$ for all smooth and irregular $3$-folds $X$ of general type over $\mathbb{C}$. For those $3$-folds $X$ attaining the equality, we completely describe their canonical models and show that the topological fundamental group $\pi_1(X) \simeq \mathbb{Z}^2$. As a

Alyssa Allende Motz, Murat Yessenov, Ayman F. Abouraddy

Space-time (ST) wave packets are propagation-invariant pulsed optical beams whose group velocity can be tuned in free space by tailoring their spatio-temporal spectral structure. To date, efforts on synthesizing ST wave packets have striven to maintain their propagation invariance. Here, we demonstrate that one degree of freedom of a ST wave packet -- its on

Liang OuYang, Dong Wang, Xiongying Qiao, Mengjie Wang

We construct a family of solutions of the holographic insulator/superconductor phase transitions with the excited states in the AdS soliton background by using both the numerical and analytical methods. The interesting point is that the improved Sturm-Liouville method can not only analytically investigate the properties of the phase transition with the excit

Hao Zheng, Jidun Wu, Qilu Cao, Jiaojiao Zhang

An 18-level argon collisional radiative model (CRM) suitable for low pressure was established. The model can be solved by combining the optical emission spectroscopy (OES) with Langmuir probe calibration. In the capacitively coupled plasmas (CCPs) with different frequency and power, the electron temperature and density obtained by the model were compared wit

Eun-Kyung Lim, Heesu Yang, Vasyl Yurchyshyn, Jongchul Chae

Light bridges (LBs) are relatively bright structures that divide sunspot umbrae into two or more parts. Chromospheric LBs are known to be associated with various activities including fan-shaped jet-like ejections and brightenings. Although magnetic reconnection is frequently suggested to be responsible for such activities, not many studies present firm evide

Setareh Rahimi Taghanaki, Michael Rainbow, Ali Etemad

We propose the use of self-supervised learning for human activity recognition with smartphone accelerometer data. Our proposed solution consists of two steps. First, the representations of unlabeled input signals are learned by training a deep convolutional neural network to predict a segment of accelerometer values. Our model exploits a novel scheme to leve

Benedito Leandro, Ana Paula de Melo, Ilton Menezes, Romildo Pina

In this paper we study the static Einstein-Maxwell space when it is conformal to an $n$-dimensional pseudo-Euclidean space, which is invariant under the action of an $(n-1)$-dimensional translation group. We also provide a complete classification of such space.

Brian P. Hanley

Four radical ideas are presented. First, that the rationale for cancellation of principal can be modified in modern banking. Second, that non-cancellation of loan principal upon payment may cure an old problem of maintenance of positive equity in the non-governmental sector. Third and fourth, that crediting this money to local/state government, and crediting

Z. Y. Zhao, G. Q. Zhang, Y. Y. Wang, Z. L. Tu

Since the discovery of FRB 200428 associated with the Galactic SGR 1935+2154, magnetars are considered to power fast radio bursts (FRBs). It is widely believed that magnetars could form by core-collapse (CC) explosions and compact binary mergers, such as binary neutron star (BNS), binary white dwarfs (BWD), and neutron star-white dwarf (NSWD) mergers. Theref

Xinyan Zhao, Liangwei Chen, Huanhuan Chen

Knowledge based dialogue systems have attracted increasing research interest in diverse applications. However, for disease diagnosis, the widely used knowledge graph is hard to represent the symptom-symptom relations and symptom-disease relations since the edges of traditional knowledge graph are unweighted. Most research on disease diagnosis dialogue system

Yong-Jiang Xu, Yong-Lu Liu, Ming-Qiu Huang

In this paper, we consider all P-wave $\Omega_{b}$ states represented by interpolating currents with a derivative and calculate the corresponding masses and pole residues with the method of QCD sum rule. Due to the large uncertainties in our calculation compared with the small difference in the masses of the excited $\Omega_{b}$ states observed by the LHCb c

Kuang-Yu Jeng, Yueh-Cheng Liu, Zhe Yu Liu, Jen-Wei Wang

We proposed an end-to-end grasp detection network, Grasp Detection Network (GDN), cooperated with a novel coarse-to-fine (C2F) grasp representation design to detect diverse and accurate 6-DoF grasps based on point clouds. Compared to previous two-stage approaches which sample and evaluate multiple grasp candidates, our architecture is at least 20 times faste

Antoine Perquin, Erica Cooper, Junichi Yamagishi

End-to-end models, particularly Tacotron-based ones, are currently a popular solution for text-to-speech synthesis. They allow the production of high-quality synthesized speech with little to no text preprocessing. Indeed, they can be trained using either graphemes or phonemes as input directly. However, in the case of grapheme inputs, little is known concer

Gábor Székelyhidi, Ben Weinkove

We consider convex solutions of nonlinear elliptic equations which satisfy the structure condition of Bian-Guan. We prove a weak Harnack inequality for the eigenvalues of the Hessian of these solutions. This can be viewed as a quantitative version of the constant rank theorem.

Ziqi Fan, Vibhav Vineet, Chenshen Lu, T. W. Wu

Acoustic scattering is strongly influenced by boundary geometry of objects over which sound scatters. The present work proposes a method to infer object geometry from scattering features by training convolutional neural networks. The training data is generated from a fast numerical solver developed on CUDA. The complete set of simulations is sampled to gener

Matthew S. Fox, Chaitanya Karamchedu

We show that the absolute convergence of a Ramanujan expansion does not guarantee the convergence of its real variable generalization, which is obtained by replacing the integer argument in the Ramanujan sums with a real number. We also construct a new Ramanujan expansion for the divisor function. While our expansion is amenable to a continuous and absolutel

Yiruo Lin

The conspiracy of ontic states responding to measurements contextually to comply with noncontextual quantum mechanical probabilities is analyzed for general ontological models. A general physical picture of ontological space structure and how ontic states are disturbed by measurement contexts is presented. A common assumption in ontological models called lam

Ahmed Hareedy, Beyza Dabak, Robert Calderbank

Constrained codes are used to prevent errors from occurring in various data storage and data transmission systems. They can help in increasing the storage density of magnetic storage devices, in managing the lifetime of electronic storage devices, and in increasing the reliability of data transmission over wires. We recently introduced families of lexicograp

Hojjat Aghakhani, Lea Schönherr, Thorsten Eisenhofer, Dorothea Kolossa

Despite remarkable improvements, automatic speech recognition is susceptible to adversarial perturbations. Compared to standard machine learning architectures, these attacks are significantly more challenging, especially since the inputs to a speech recognition system are time series that contain both acoustic and linguistic properties of speech. Extracting

Vishal Mandal, Abdul Rashid Mussah, Yaw Adu-Gyamfi

Automatic detection and classification of pavement distresses is critical in timely maintaining and rehabilitating pavement surfaces. With the evolution of deep learning and high performance computing, the feasibility of vision-based pavement defect assessments has significantly improved. In this study, the authors deploy state-of-the-art deep learning algor

Mingshang Hu, Shaolin Ji, Rundong Xu

We study a stochastic optimal control problem for forward-backward control systems with quadratic generators. In order to establish the first and second-order variational and adjoint equations, we obtain a new estimate for one-dimensional linear BSDEs with unbounded stochastic Lipschitz coefficients involving bounded mean oscillation martingales (BMO-marting

Rodrigo F. Neumann, Mariane Barsi-Andreeta, Everton Lucas-Oliveira, Hugo Barbalho

Permeability is the key parameter for quantifying fluid flow in porous rocks. Knowledge of the spatial distribution of the connected pore space allows, in principle, to predict the permeability of a rock sample. However, limitations in feature resolution and approximations at microscopic scales have so far precluded systematic upscaling of permeability predi

Layton A. Hall, Ayman F. Abouraddy

'Space-time' (ST) wave packets are propagation-invariant pulsed optical beams that travel rigidly in linear media without diffraction or dispersion at a potentially arbitrary group velocity. These unique characteristics are a result of spatio-temporal spectral correlations introduced into the field; specifically, each spatial frequency is associated with a s

Yunpeng Li, Marco Tagliasacchi, Oleg Rybakov, Victor Ungureanu

In this paper we propose a lightweight model for frequency bandwidth extension of speech signals, increasing the sampling frequency from 8kHz to 16kHz while restoring the high frequency content to a level almost indistinguishable from the 16kHz ground truth. The model architecture is based on SEANet (Sound EnhAncement Network), a wave-to-wave fully convoluti

Ismail Laraiedh

In this paper we define and discuss the representations of $n$-BiHom-Lie algebra. We also introduce $T_θ$-extensions and $T_θ^{\ast}$-extensions of $n$-BiHom-Lie algebras and prove the necessary and sufficient conditions for a $2m$-dimensional quadratic $n$-Bihom-Lie algebra to be isomorphic to a $T_θ^{\ast}$-extension. Moreover, we develop the one-parameter

N. Keeley, K. W. Kemper, K. Rusek

While it is well established that the ground state reorientation coupling can have a significant influence on the elastic scattering of deformed nuclei, the effect of such couplings on transfer channels has been much less well investigated. In this letter we demonstrate that the 208Pb(7Li,6He)209Bi proton stripping reaction at an incident energy of 52 MeV ca

Alexander William Wong, Weijie Sun, Sunil Vasu Kalmady, Padma Kaul

The 12-lead electrocardiogram (ECG) is a commonly used tool for detecting cardiac abnormalities such as atrial fibrillation, blocks, and irregular complexes. For the PhysioNet/CinC 2020 Challenge, we built an algorithm using gradient boosted tree ensembles fitted on morphology and signal processing features to classify ECG diagnosis. For each lead, we derive

Aaqib Saeed, David Grangier, Olivier Pietquin, Neil Zeghidour

We propose CHARM, a method for training a single neural network across inconsistent input channels. Our work is motivated by Electroencephalography (EEG), where data collection protocols from different headsets result in varying channel ordering and number, which limits the feasibility of transferring trained systems across datasets. Our approach builds upon

Daniel C. Mayer

For each odd prime p>=5, there exist finite p-groups G with derived quotient G/D(G)=C(p)xC(p) and nearly constant transfer kernel type k(G)=(1,2,...,2) having two fixed points. It is proved that, for p=7, this type k(G) with the simplest possible case of logarithmic abelian quotient invariants t(G)=(11111,111,21,21,21,21,21,21) of the eight maximal subgroups

Xiaofeng Liu, Yuzhuo Han, Song Bai, Yi Ge

Semantic segmentation (SS) is an important perception manner for self-driving cars and robotics, which classifies each pixel into a pre-determined class. The widely-used cross entropy (CE) loss-based deep networks has achieved significant progress w.r.t. the mean Intersection-over Union (mIoU). However, the cross entropy loss can not take the different impor

Jun-Hui Zhao, Mark R. Morris, W. M. Goss

The radio bright zone (RBZ) at the Galactic center has been observed with the JVLA in the A, B and C array configurations at 5.5 and 9 GHz. With a procedure for high-dynamic range imaging developed on CASA, we constructed deep images a resolution up to 0.2", achieving rms noises of a few $μ$Jy/beam. From the high-resolution and high-dynamics range images

Upasak Das, Prasenjit Sarkhel, Sania Ashraf

Compliance with measures like social distancing, hand-washing and wearing masks have emerged as the dominant strategy to combat health risk from the COVID-19 pandemic. These behaviors are often argued to be pro-social, where one must incur private cost to benefit or protect others. Using self-reported data across India (n=934) through online survey, we asses

Goodarz Mehr

This paper introduces MuRAD (Musculoskeletal Radiograph Abnormality Detection tool), a tool that can help radiologists automate the detection of abnormalities in musculoskeletal radiographs (bone X-rays). MuRAD utilizes a Convolutional Neural Network (CNN) that can accurately predict whether a bone X-ray is abnormal, and leverages Class Activation Map (CAM)

Patrick Das Gupta

Beginning with the basic notions of quantum theory, impossibility of `trajectory' description for particles that ensues from uncertainty principle is discussed. Why the observed tracks in bubble/cloud chambers are not really the `trajectories' of high energy particles, rather they are simply the trails of the atoms/molecules excited or ionized in the

Lizaveta Miasayedava, Keegan McBride, Jeffrey Andrew Tuhtan

Negative environmental impacts on societies and ecosystems are frequently driven by human activity and amplified by increasing climatic variability. Properly managing these impacts relies on a government's ability to ensure environmental regulatory compliance in the face of increasing uncertainty. Water flow rates are the most widely used evaluation metr

David Terman, Yousef Hannawi

Maintaining cerebral blood flow is critical for adequate neuronal function. Previous computational models of brain capillary networks have predicted that heterogeneous cerebral capillary flow patterns result in lower brain tissue partial oxygen pressures. It has been suggested that this may lead to number of diseases such as Alzheimer's disease, acute is

Luis Sequeira, Toktam Mahmoodi

In this paper, we study the role that machine learning can play in cooperative driving. Given the increasing rate of connectivity in modern vehicles, and road infrastructure, cooperative driving is a promising first step in automated driving. The example scenario we explored in this paper, is coordinated lane merge, with data collection, test and evaluation

Jiancheng Yang, Yangzhou Jiang, Xiaoyang Huang, Bingbing Ni

This paper addresses the challenging black-box adversarial attack problem, where only classification confidence of a victim model is available. Inspired by consistency of visual saliency between different vision models, a surrogate model is expected to improve the attack performance via transferability. By combining transferability-based and query-based blac

Cheng Chen, Luo Luo, Weinan Zhang, Yong Yu

The Frank-Wolfe algorithm is a classic method for constrained optimization problems. It has recently been popular in many machine learning applications because its projection-free property leads to more efficient iterations. In this paper, we study projection-free algorithms for convex-strongly-concave saddle point problems with complicated constraints. Our

Betania S. C. Campello, Leonardo T. Duarte, João M. T. Romano

A number of Multiple Criteria Decision Analysis (MCDA) methods have been developed to rank alternatives based on several decision criteria. Usually, MCDA methods deal with the criteria value at the time the decision is made without considering their evolution over time. However, it may be relevant to consider the criteria' time series since providing ess

An Nguyen, Wenyu Zhang, Leo Schwinn, Bjoern Eskofier

Process Mining has recently gained popularity in healthcare due to its potential to provide a transparent, objective and data-based view on processes. Conformance checking is a sub-discipline of process mining that has the potential to answer how the actual process executions deviate from existing guidelines. In this work, we analyze a medical training proce

Vyacheslav V. Chistyakov, Svetlana A. Chistyakova

Let $T\subset\mathbb{R}$ and $(X,\mathcal{U})$ be a uniform space with an at most countable gage of pseudometrics $\{d_p:p\in\mathcal{P}\}$ of the uniformity $\mathcal{U}$. Given $f\in X^T$ (=the family of all functions from $T$ into $X$), the approximate variation of $f$ is the two-parameter family $\{V_{\varepsilon,p}(f):\varepsilon>0,p\in\mathcal{P}\}$, w

M. Li, H. Bai, L. Tan, K. Xiong

Measuring sentence semantic similarity using pre-trained language models such as BERT generally yields unsatisfactory zero-shot performance, and one main reason is ineffective token aggregation methods such as mean pooling. In this paper, we demonstrate under a Bayesian framework that distance between primitive statistics such as the mean of word embeddings

Sam Spencer, Lav R. Varshney

Previous work has shown that for contagion processes on extended star networks (trees with exactly one node of degree > 2), there is a simple, closed-form expression for a highly accurate approximation to the maximum likelihood infection source. Here, we generalize that result to a class of hypertrees which, although somewhat structurally analogous, provides

Xie Chen, Sarangarajan Parthasarathy, William Gale, Shuangyu Chang

LSTM language models (LSTM-LMs) have been proven to be powerful and yielded significant performance improvements over count based n-gram LMs in modern speech recognition systems. Due to its infinite history states and computational load, most previous studies focus on applying LSTM-LMs in the second-pass for rescoring purpose. Recent work shows that it is fe

Davor Horvatic, Dubravko Klabucar, Dalibor Kekez

Exploring the extent of model dependence, we study effects of certain Ansaetze for the T-dependence of the correction term in the QCD topological susceptibility. The one producing unwanted effects on results at T > 0 in the eta'-eta complex in the usual limit of isospin symmetry, is largely cured from its peculiar behavior and brought into agreement with

Chiraag Kaushik, T. Mitchell Roddenberry, Santiago Segarra

We consider the problem of sequential graph topology change-point detection from graph signals. We assume that signals on the nodes of the graph are regularized by the underlying graph structure via a graph filtering model, which we then leverage to distill the graph topology change-point detection problem to a subspace detection problem. We demonstrate how

Brian G. O'Flynn, Tanja Mittag

Liquid-liquid phase separation is now recognized as a common mechanism for regulating enzyme activity in cells. Insights from studies in cells are complemented by in vitro studies aimed at developing better understanding of mechanisms underlying such control. These mechanisms are often based on the influence of LLPS on the physicochemical properties of the e

Soroosh Yazdani, Andrea Tagliasacchi

In this technical report, we investigate extending convolutional neural networks to the setting where functions are not sampled in a grid pattern. We show that by treating the samples as the average of a function within a cell, we can find a natural equivalent of most layers used in CNN. We also present an algorithm for running inference for these models exa

Samuel Faucher, Daniel James Lundberg, Xinyao Anna Liang, Xiaojia Jin

While facial coverings over the nose and mouth reduce the spread of the virus SARS-CoV-2 by filtration, masks capable of viral inactivation by heating could provide a complementary method to limit viral transmission. In this work, we introduce a new virucidal face mask concept based on a reverse-flow reactor driven by the oscillatory flow of human breath. Th

T. B. Issa, R. B Salako, W. Shen

In this paper, we consider two species chemotaxis systems with Lotka-Volterra competition reaction terms. Under appropriate conditions on the parameters in such a system, we establish the existence of traveling wave solutions of the system connecting two spatially homogeneous equilibrium solutions with wave speed greater than some critical number c*. We also

Yogarshi Vyas, Miguel Ballesteros

In entity linking, mentions of named entities in raw text are disambiguated against a knowledge base (KB). This work focuses on linking to unseen KBs that do not have training data and whose schema is unknown during training. Our approach relies on methods to flexibly convert entities from arbitrary KBs with several attribute-value pairs into flat strings, w

David Arbour, Drew Dimmery, Anup Rao

In this work, we reframe the problem of balanced treatment assignment as optimization of a two-sample test between test and control units. Using this lens we provide an assignment algorithm that is optimal with respect to the minimum spanning tree test of Friedman and Rafsky (1979). This assignment to treatment groups may be performed exactly in polynomial t

Jesse Railo

This PhD dissertation is concerned with integral geometric inverse problems. The geodesic ray transform is an operator that encodes the line integrals of a function along geodesics. The dissertation establishes many conditions when such information determines a function uniquely and stably. A new numerical model for computed tomography imaging is created as

Rachel C. Nethery, Nina Katz-Christy, Marianthi-Anna Kioumourtzoglou, Robbie M. Parks

Strategic preparedness has been shown to reduce the adverse health impacts of hurricanes and tropical storms, referred to collectively as tropical cyclones (TCs), but its protective impact could be enhanced by a more comprehensive and rigorous characterization of TC epidemiology. To generate the insights and tools necessary for high-precision TC preparedness

Ashley Cronk, David Larson, Francesca M. Toma

A continuous CO2 vapor-fed electrochemical cell prototype that performs CO2R with high faradaic efficiency for desired carbon products for up to 72 hours of operation is presented. The cell design facilitates a flow through configuration, capitalizing on gas diffusion electrode (GDE) and membrane electrode architecture (MEA) without requiring a catholyte. We

Dhananjay Ashok, Joseph Scott, Sebastian Wetzel, Maysum Panju

We present a novel Auxiliary Truth enhanced Genetic Algorithm (GA) that uses logical or mathematical constraints as a means of data augmentation as well as to compute loss (in conjunction with the traditional MSE), with the aim of increasing both data efficiency and accuracy of symbolic regression (SR) algorithms. Our method, logic-guided genetic algorithm (

Rui Feng, Jie Yuan, Chao Zhang

We study the problem of event extraction from text data, which requires both detecting target event types and their arguments. Typically, both the event detection and argument detection subtasks are formulated as supervised sequence labeling problems. We argue that the event extraction models so trained are inherently label-hungry, and can generalize poorly

Tin Lai, Fabio Ramos

Sampling-based motion planning is the predominant paradigm in many real-world robotic applications, but its performance is immensely dependent on the quality of the samples. The majority of traditional planners are inefficient as they use uninformative sampling distributions as opposed to exploiting structures and patterns in the problem to guide better samp

Jaehong Park, Jonathan Pilault, Christopher Pal

We introduce Mem2Mem, a memory-to-memory mechanism for hierarchical recurrent neural network based encoder decoder architectures and we explore its use for abstractive document summarization. Mem2Mem transfers "memories" via readable/writable external memory modules that augment both the encoder and decoder. Our memory regularization compresses an en

Subrata Sarkar, Rizwan Ahmad, Philip Schniter

Motivated by image recovery in magnetic resonance imaging (MRI), we propose a new approach to solving linear inverse problems based on iteratively calling a deep neural-network, sometimes referred to as plug-and-play recovery. Our approach is based on the vector approximate message passing (VAMP) algorithm, which is known for mean-squared error (MSE)-optimal

Krzysztof Burkat, Maciej Pawlik, Bartosz Balis, Maciej Malawski

Increasing popularity of the serverless computing approach has led to the emergence of new cloud infrastructures working in Container-as-a-Service (CaaS) model like AWS Fargate, Google Cloud Run, or Azure Container Instances. They introduce an innovative approach to running cloud containers where developers are freed from managing underlying resources. In th

M. Scott Osborne, Garth Warner

This paper initiates a study into the contribution to the trace provided by the conjugacy classes.

José Mairton B. da Silva, Gustav Wikström, Ratheesh K. Mungara, Carlo Fischione

Although in cellular networks full-duplex and dynamic time-division duplexing promise increased spectrum efficiency, their potential is so far challenged by increased interference. While previous studies have shown that self-interference can be suppressed to a sufficient level, we show that the cross-link interference for both duplexing modes, especially fro

Rhys T. Bury, Alexander V. Mikhailov

We study automorphic Lie algebras and their applications to integrable systems. Automorphic Lie algebras are a natural generalisation of celebrated Kac-Moody algebras to the case when the group of automorphisms is not cyclic. They are infinite dimensional and almost graded. We formulate the concept of a graded isomorphism and classify $sl(2,C)$ based automor

Maximillian Hart, Mark G. Kuzyk

We use videos taken with a mobile phone to study conservation of energy, conservation of momentum, and the work-energy theorem by analyzing the collision of a cue ball and the eight ball. A video of the full time sequence, starting from before the cue ball is struck until well after the collision, is recorded with a mobile phone. The video is imported into O

Christopher C. Finlay, Clemens Kloss, Nils Olsen, Magnus D. Hammer

We present the CHAOS-7 model of the time-dependent near-Earth geomagnetic field between 1999 and 2020 based on magnetic field observations collected by the low-Earth orbit satellites {\it Swarm}, CryoSat-2, CHAMP, SAC-C and Ørsted, and on annual differences of monthly means of ground observatory measurements. The CHAOS-7 model consists of a time-dependent in

Assim Boukhayma

This paper focuses on the conversion gain (CG) of pixels implementing pinned photo-diodes (PPD) and in-pixel voltage follower in standard CMOS image sensor (CIS) process. An overview of the CG expression and its impact on the noise performance of the CIS readout chain is presented. CG enhancement techniques involving process refinements and pure circuit desi

Ayyoub Ahar, Tobias Birnbaum, Maksymilian Chlipala, Weronika Zaperty

Objective quality assessment of digital holograms has proven to be a challenging task. While prediction of perceptual quality of the recorded 3D content from the holographic wavefield is an open problem; perceptual quality assessment from content after rendering, requires a time-consuming rendering step and a multitude of possible viewports. In this research

Xueru Zhang, Ruibo Tu, Yang Liu, Mingyan Liu

Although many fairness criteria have been proposed for decision making, their long-term impact on the well-being of a population remains unclear. In this work, we study the dynamics of population qualification and algorithmic decisions under a partially observed Markov decision problem setting. By characterizing the equilibrium of such dynamics, we analyze t

C Marques, G S Dias, H S Chavez, S B Duarte

The Mössbauer spectroscopy is presented as an alternative experimental technique to be pursued in the detec-tion of Coherent Elasticν-Nucleus Scattering (CENNS). The neutrino transferred energy in the neutrino-nucleusinteraction causes a perturbation at the nuclear levels which are responsible for gamma radiation in Mössbauerresonance. The main characteristi

Drew Lilley, Jonathan Lau, Chris Dames, Sumanjeet Kaur

Phase change material based thermal energy storage has many current and potential applications in the heating and cooling of buildings, battery and electronics thermal management, thermal textiles, and dry cooling of power plants. However, connecting lab scale thermal data obtained on DSC to the performance of large-scale practical systems has been a major c

Halima Bouzidi, Hamza Ouarnoughi, Smail Niar, Abdessamad Ait El Cadi

Running Convolutional Neural Network (CNN) based applications on edge devices near the source of data can meet the latency and privacy challenges. However due to their reduced computing resources and their energy constraints, these edge devices can hardly satisfy CNN needs in processing and data storage. For these platforms, choosing the CNN with the best tr