May 2022 arXiv papers — page 51

Yuan Yao, Qianyu Chen, Ao Zhang, Wei Ji

Vision-language pre-training (VLP) has shown impressive performance on a wide range of cross-modal tasks, where VLP models without reliance on object detectors are becoming the mainstream due to their superior computation efficiency and competitive performance. However, the removal of object detectors also deprives the capability of VLP models in explicit ob

Xuefeng Gao, Xun Yu Zhou

We consider reinforcement learning for continuous-time Markov decision processes (MDPs) in the infinite-horizon, average-reward setting. In contrast to discrete-time MDPs, a continuous-time process moves to a state and stays there for a random holding time after an action is taken. With unknown transition probabilities and rates of exponential holding times,

Jiming Ma, Baohua Xie

We show the manifolds at infinity of the complex hyperbolic triangle groups $\Delta_{3,4,4;\infty}$ and $\Delta_{3,4,6;\infty}$,are one-cusped hyperbolic 3-manifolds $m038$ and $s090$ in the Snappy Census respectively.That is,these two manifolds admit spherical CR uniformizations. Moreover, these two hyperbolic 3-manifolds above can be obtained by Dehn surge

Yuan Yao, Bowen Dong, Ao Zhang, Zhengyan Zhang

Recent works have shown promising results of prompt tuning in stimulating pre-trained language models (PLMs) for natural language processing (NLP) tasks. However, to the best of our knowledge, existing works focus on prompt-tuning generative PLMs that are pre-trained to generate target tokens, such as BERT. It is still unknown whether and how discriminative

Heesang Park, Dongsoo Shin

We investigate the correspondence between three theories of deformations of rational surface singularities: de Jong and van Straten's picture deformations, Koll\'ar's P-resolutions, and Pinkham's smoothings of negative weights. We provide an explicit method for obtaining, from a given deformation in one theory, deformations in other theories that parameteriz

Kazuyasu Shigemoto

In the previous paper, we reviewed the Rosenhain's paper to the Jacobi's inversion problem for the genus two hyperelliptic integral. In this paper, we review the G\"{o}pel's paper to the Jacobi's inversion problem for the genus two hyperelliptic integral.

Padmanaba Srinivasan, William J. Knottenbelt

Many real-world tasks are plagued by limitations on data: in some instances very little data is available and in others, data is protected by privacy enforcing regulations (e.g. GDPR). We consider limitations posed specifically on time-series data and present a model that can generate synthetic time-series which can be used in place of real data. A model tha

Víctor Medina

This article presents some aspects and experience in the use of algebraic manipulation software applied to general relativity. Some years ago certain results were reported using computer algebra platforms, but the growing popularity of graphical platforms such as Maple allows us to approach the problem of the simplifications of many expressions from another

Yue Jin, Shuangqing Wei, Jian Yuan, Xudong Zhang

Learning to coordinate is a daunting problem in multi-agent reinforcement learning (MARL). Previous works have explored it from many facets, including cognition between agents, credit assignment, communication, expert demonstration, etc. However, less attention were paid to agents' decision structure and the hierarchy of coordination. In this paper, we explo

Weizhen Qi, Yeyun Gong, Yelong Shen, Jian Jiao

Non-Autoregressive generation is a sequence generation paradigm, which removes the dependency between target tokens. It could efficiently reduce the text generation latency with parallel decoding in place of token-by-token sequential decoding. However, due to the known multi-modality problem, Non-Autoregressive (NAR) models significantly under-perform Auto-r

Ang Li, Yiming Pan, Philip Dienstbier, Peter Hommelhoff

When two-color femtosecond laser pulses interact with matter, electrons can be emitted through various multiphoton excitation pathways. Quantum interference between these pathways gives rise to a strong oscillation of the photoemitted electron current, experimentally characterized by its visibility. In this work, we demonstrate two-color visibility spectrosc

Yoshiaki Tsujimoto, Rikizo Ikuta, Kentaro Wakui, Toshiki Kobayashi

We propose a method to perform the quantum state tomography (QST) of an $\mathit{n}$-partite qudit state embedded in single photons using the Hong-Ou-Mandel (HOM) interference between the target state and probe state. This method requires only passive beam splitters for the HOM interference and removes all active optical devices in the target modes to contro

Yuki Takezawa, Kenta Niwa, Makoto Yamada

In recent years, decentralized learning has emerged as a powerful tool not only for large-scale machine learning, but also for preserving privacy. One of the key challenges in decentralized learning is that the data distribution held by each node is statistically heterogeneous. To address this challenge, the primal-dual algorithm called the Edge-Consensus Le

Ekaterina Artemova, Maxim Zmeev, Natalia Loukachevitch, Igor Rozhkov

The RuNNE Shared Task approaches the problem of nested named entity recognition. The annotation schema is designed in such a way, that an entity may partially overlap or even be nested into another entity. This way, the named entity "The Yermolova Theatre" of type "organization" houses another entity "Yermolova" of type "person". We adopt the Russian NEREL d

Jie Zhang, Chen Chen, Lingjuan Lyu

Knowledge Distillation (KD) is a typical method for training a lightweight student model with the help of a well-trained teacher model. However, most KD methods require access to either the teacher's training data or model parameters, which is unrealistic. To tackle this problem, recent works study KD under data-free and black-box settings. Nevertheless, the

Aditya Jain, Amin A. Nizami

We construct superconformal invariants in superspace which are used to build 3-point correlators of spinning operators in general $\cal{N}=2$ superconformal field theories in three dimensions. Our systematic analysis includes various relations between these invariants and provides a minimal set of parity-even and parity-odd invariants which is further used t

Qizhang Li, Yiwen Guo, Wangmeng Zuo, Hao Chen

The vulnerability of deep neural networks (DNNs) to adversarial examples has attracted great attention in the machine learning community. The problem is related to non-flatness and non-smoothness of normally obtained loss landscapes. Training augmented with adversarial examples (a.k.a., adversarial training) is considered as an effective remedy. In this pape

Yuki Miyazaki, Giacomo Marmorini, Nobuo Furukawa, Daisuke Yamamoto

We study the quantum magnetism of the SU(4) Mott insulator in a square optical superlattice, in which atoms with four nuclear-spin components strongly interact with each other, in the presence of an external field that controls the imbalance between the population of two components and that of the other two. This is a natural extension of the physics of spin

Hamed Masoumi, Nitin Jonathan Myers, Geert Leus, Sander Wahls

Fast millimeter wave (mmWave) channel estimation techniques based on compressed sensing (CS) suffer from low signal-to-noise ratio (SNR) in the channel measurements, due to the use of wide beams. To address this problem, we develop an in-sector CS-based mmWave channel estimation technique that focuses energy on a sector in the angle domain. Specifically, we

Jan Decuyper, Koen Tiels, Siep Weiland, Mark C. Runacres

Multivariate functions emerge naturally in a wide variety of data-driven models. Popular choices are expressions in the form of basis expansions or neural networks. While highly effective, the resulting functions tend to be hard to interpret, in part because of the large number of required parameters. Decoupling techniques aim at providing an alternative rep

Daisuke Kikuta, Toyotaro Suzumura, Md Mostafizur Rahman, Yu Hirate

Leveraging graphs on recommender systems has gained popularity with the development of graph representation learning (GRL). In particular, knowledge graph embedding (KGE) and graph neural networks (GNNs) are representative GRL approaches, which have achieved the state-of-the-art performance on several recommendation tasks. Furthermore, combination of KGE and

Euro Spallucci, Anais Smailagic

Similarity in the structure of scattering amplitudes in Yang-Mills theories and General Relativity led to the idea that graviton could be described as the double copy of a vector gauge field. In this letter we discuss a realization of this idea emerging directly from solutions of equations of General Relativity. A general form of the energy momentum tensor f

Meryem M'hamdi, Xiang Ren, Jonathan May

The longstanding goal of multi-lingual learning has been to develop a universal cross-lingual model that can withstand the changes in multi-lingual data distributions. There has been a large amount of work to adapt such multi-lingual models to unseen target languages. However, the majority of work in this direction focuses on the standard one-hop transfer le

David Yallup, Will Handley, Mike Hobson, Anthony Lasenby

The true posterior distribution of a Bayesian neural network is massively multimodal. Whilst most of these modes are functionally equivalent, we demonstrate that there remains a level of real multimodality that manifests in even the simplest neural network setups. It is only by fully marginalising over all posterior modes, using appropriate Bayesian sampling

Ze-Hua Cao, Wei He, Zhi-Feng Sun

In this work, we construct the Lagrangian describing meson-diquark interaction, such that the diquark-anti-diquark component as well as the molecular component is introduced when studying the $Z_{cs}$ states. In this way, the problem is solved that if only considering the $\bar{D}^{(*)}D_s^{(*)}$ components, the potentials are suppressed by OZI rule. Through

Tom Gustafsson, Philip L. Lederer

We consider mixed finite element approximations of viscous, plastic Bingham flow in a cylindrical pipe. A novel a priori and a posteriori error analysis is introduced which is based on a discrete mesh dependent norm for the normalized Lagrange multiplier. This allows proving stability for various conforming finite elements. Numerical examples are presented t

Keren Censor-Hillel, Shir Cohen, Ran Gelles, Gal Sela

We address fully-defective asynchronous networks, in which all links are subject to an unlimited number of alteration errors, implying that all messages in the network may be completely corrupted. Despite the possible intuition that such a setting is too harsh for any reliable communication, we show how to simulate any algorithm for a noiseless setting over

Sonali Kaushik, Rajesh Kumar

The article presents the existence and mass conservation of solution for the discrete Safronov-Dubovski coagulation equation for the product coalescence coefficients $\phi$ such that $\phi_{i,j} \leq ij$ $\forall$ $i,j \in \mathbb{N}$. Both conservative and non-conservative truncated systems are used to analyse the infinite system of ODEs. In the conservativ

Félix Lequen

We prove that an homogeneous space of an almost simple Lie group does not have any measurable tiling invariant by a lattice of the Lie group. This refines the Howe-Moore ergodicity theorem. -- On d\'emontre qu'un espace homog\`ene d'un groupe de Lie presque simple n'admet pas de pavage mesurable invariant par un r\'eseau du groupe de Lie. Ceci constitue un r

Wei Wang, Xin Zhang, Jiaqi Yi, Xianqi Liao

This study aimed to improve the surface defect detection accuracy of ZrO2 ceramic bearing balls. Combined with the noise damage of the image samples, a surface defect detection method for ZrO2 ceramic bearing balls based on cartoon-texture decomposition model was proposed. Building a ZrO2 ceramic bearing ball surface defect detection system. The ZrO2 ceramic

Joel C. F. Balzan, Miroslav D. Filipovic, Shi Dai, Rami Z. E. Alsaberi

We present radio continuum observations of NGC 2082 using ASKAP, ATCA and Parkes telescopes from 888 MHz to 9000 MHz. Some 20 arcsec from the centre of this nearby spiral galaxy, we discovered a bright and compact radio source, J054149.24-641813.7, of unknown origin. To constrain the nature of J054149.24-641813.7, we searched for transient events with the Ul

S. L. West, R. Lunnan, C. M. B. Omand, T. Kangas

SN\,2020qlb (ZTF20abobpcb) is a hydrogen-poor superluminous supernova (SLSN-I) that is among the most luminous (maximum M$_{g} = -22.25$ mag) and that has one of the longest rise times (77 days from explosion to maximum). We estimate the total radiated energy to be $>2.1\times10^{51}$ erg. SN\,2020qlb has a well-sampled light curve that exhibits clear near a

Fabio Nicola, S. Ivan Trapasso

Within the mathematical analysis of deep convolutional neural networks, the wavelet scattering transform introduced by St\'ephane Mallat is a unique example of how the ideas of multiscale analysis can be combined with a cascade of modulus nonlinearities to build a nonexpansive, translation invariant signal representation with provable geometric stability pro

Giuseppe Dattoli, Silvia Licciardi

In this article we go deeply into the formulation and meaning of the monomiality principle and employ it to study the properties of a set of polynomials, which, asymptotically, reduce to the ordinary two variable Kampe de Feriet family. We derive the relevant differential equations and discuss the associated orthogonality properties, along with the relevant

Peng Hu, Xi Peng, Hongyuan Zhu, Mohamed M. Sabry Aly

As Deep Neural Networks (DNNs) usually are overparameterized and have millions of weight parameters, it is challenging to deploy these large DNN models on resource-constrained hardware platforms, e.g., smartphones. Numerous network compression methods such as pruning and quantization are proposed to reduce the model size significantly, of which the key is to

Xiaoyu Chen, Han Zhong, Zhuoran Yang, Zhaoran Wang

We study human-in-the-loop reinforcement learning (RL) with trajectory preferences, where instead of receiving a numeric reward at each step, the agent only receives preferences over trajectory pairs from a human overseer. The goal of the agent is to learn the optimal policy which is most preferred by the human overseer. Despite the empirical successes, the

Xu Chen, Qiu Qiu, Changshan Li, Kunqing Xie

In recent years, the emergence and development of third-party platforms have greatly facilitated the growth of the Online to Offline (O2O) business. However, the large amount of transaction data raises new challenges for retailers, especially anomaly detection in operating conditions. Thus, platforms begin to develop intelligent business assistants with embe

Félix Lequen

In this note, we generalise a Bourgain's construction of finitely-supported symmetric measures whose Furstenberg measure has a smooth density from the case of $\mathrm{SL}_2(\mathbb{R})$ to that of a general simple Lie group. The proof is the same as Bourgain's, except that the use of Fourier series is replaced by harmonic analysis on a maximal compact subgr

Yang ChaoQun

Machine learning algorithms are undoubtedly one of the most popular algorithms in recent years, and neural networks have demonstrated unprecedented precision. In daily life, different communities may have different user characteristics, which also means that training a strong model requires the union of different communities, so the privacy issue needs to be

U. Simola, A. Bonfanti, X. Dumusque, J. Cisewski-Kehe

Active regions on the photosphere of a star have been the major obstacle for detecting Earth-like exoplanets using the radial velocity (RV) method. A commonly employed solution for addressing stellar activity is to assume a linear relationship between the RV observations and the activity indicators along the entire time series, and then remove the estimated

Baihong Li, Boxin Yuan, Changhua Chen, Xiao Xiang

A modified Hong--Ou--Mandel (HOM) interference reveals that the two-photon interference phenomenon can be explained only by the concept of a two-photon wave packet rather than a single-photon one. However, the temporal interferogram in the modified HOM interferometer becomes flat in some cases so that no useful information can be extracted from time-domain m

Yves Bestgen

This paper argues for the widest possible use of bootstrap confidence intervals for comparing NLP system performances instead of the state-of-the-art status (SOTA) and statistical significance testing. Their main benefits are to draw attention to the difference in performance between two systems and to help assessing the degree of superiority of one system o

Kenji Itao, Kunihiko Kaneko

Anthropologists have observed gift relationships that establish social relations as well as the transference of goods in many human societies. The totality of such social relations constitutes the network. Social scientists have analysed different types of social organisations with their characteristic networks. However, the factors and mechanisms that cause

Steffen Zelzer, Rajib Batabyal, Derek Dardzinski, Noa Marom

We combined in-situ scanning tunneling microscopy (STM) with the conventional growth characterization methods of atomic force microscopy (AFM) and reflection high energy electron diffraction (RHEED) to simultaneously assess atomic-scale impurities and the larger-scale surface morphology of molecular beam epitaxy (MBE) grown homoepitaxial InAs(111)A. By keepi

Tianshui Chen, Tao Pu, Lingbo Liu, Yukai Shi

Multi-label image recognition with partial labels (MLR-PL), in which some labels are known while others are unknown for each image, may greatly reduce the cost of annotation and thus facilitate large-scale MLR. We find that strong semantic correlations exist within each image and across different images, and these correlations can help transfer the knowledge

Benoit Clément, Stefan Baeßler, Valery V. Nesvizhevsky, Emily Perry

The bouncing neutron is one of the rare system where gravity can be studied in a quantum framework. To this end it is crucial to be able to select some specific gravitational quantum state (GQS). The GRANIT apparatus is the first physics experiment connected to a superthermal helium UCN source. We report on the methods developed for this instrument showing h

Rong-Hua Wang, Michael X. X. Zhong

Polynomial reduction, designed first for hypergeometric terms, can be used to automatically prove and generate new hypergeometric identities from old ones. In this paper, we extend the reduction method to holonomic sequences. As applications, we describe an algorithmic way to prove and generate new multi-summation identities. Especially we present new famili

F. Mayet, M. Hachmann, K. Floettmann, F. Burkart

The transverse emittance of a charged particle beam is an important figure of merit for many accelerator applications, such as ultra-fast electron diffraction, free electron lasers and the operation of new compact accelerator concepts in general. One of the easiest to implement methods to determine the transverse emittance is the phase advance scan method us

Yashar Deldjoo, Dietmar Jannach, Alejandro Bellogin, Alessandro Difonzo

Recommender systems can strongly influence which information we see online, e.g., on social media, and thus impact our beliefs, decisions, and actions. At the same time, these systems can create substantial business value for different stakeholders. Given the growing potential impact of such AI-based systems on individuals, organizations, and society, questi

Yingying Zhang, Qiaoyong Zhong, Di Xie, Shiliang Pu

The knowledge replay technique has been widely used in many tasks such as continual learning and continuous domain adaptation. The key lies in how to effectively encode the knowledge extracted from previous data and replay them during current training procedure. A simple yet effective model to achieve knowledge replay is autoencoder. However, the number of s

Felix Bock, Rafał Kalinowski, Johannes Pardey, Monika Pilśniak

We propose the notion of a majority $k$-edge-coloring of a graph $G$, which is an edge-coloring of $G$ with $k$ colors such that, for every vertex $u$ of $G$, at most half the edges of $G$ incident with $u$ have the same color. We show the best possible results that every graph of minimum degree at least $2$ has a majority $4$-edge-coloring, and that every g

Arul Selvam Periyasamy, Catherine Capellen, Max Schwarz, Sven Behnke

Object pose estimation is a key perceptual capability in robotics. We propose a fully-convolutional extension of the PoseCNN method, which densely predicts object translations and orientations. This has several advantages such as improving the spatial resolution of the orientation predictions -- useful in highly-cluttered arrangements, significant reduction

Siddhartha Patra, Anirban Mukherjee, Siddhartha Lal

An outstanding challenge involves understanding the many-particle entanglement of liquid states of quantum matter that arise in systems of interacting electrons. The Fermi liquid (FL) in $D$ spatial dimensions shows a violation of the area-law in real-space entanglement entropy of a subsystem (of length $L$), $S_{EE} \sim L^{D-1}\ln L$, widely believed to be

Y. Chang, C. Lizardi, R. Shah

Momentum and mean reversion trading strategies have opposite characteristics. The former is generally better with trending assets, and the latter is generally better with mean reverting assets. Using the Hurst exponent, which classifies time series as trending or mean reverting, we attempt to trade with each strategy when it is advantageous to generate highe

Thomas Carette

In the recent years, Local Differential Privacy (LDP) has been one of the corner stone of privacy preserving data analysis. However, many challenges still opposes its widespread application. One of these problems is the scalability of LDP to high dimensional data, in particular for estimating joint-distributions. In this paper, we develop an approximate esti

Nicholas Cazet

Yoshikawa made a table of knotted surfaces in R^4 with ch-index 10 or less. This remarkable table is the first to enumerate knotted surfaces analogous to the classical prime knot table. A broken sheet diagram of a surface-link is a generic projection of the surface in R^3 with crossing information along its singular set. The minimal number of triple points a

Jingwang Li, Housheng Su

This work focuses on a class of general decentralized constraint-coupled optimization problems. We propose a novel nested primal-dual gradient algorithm (NPGA), which can achieve linear convergence under the weakest known condition, and its theoretical convergence rate surpasses all known results. More importantly, NPGA serves not only as an algorithm but al

Gian Maria Dall'Ara, Alessandro Monguzzi

In this work we explore the theme of $L^p$-boundedness of Bergman projections of domains that can be covered, in the sense of ramified coverings, by "nice" domains (e.g. strictly pseudoconvex domains with real analytic boundary). In particular, we focus on two-dimensional normal ramified coverings whose covering group is a finite unitary reflection group. In

Paul Scherer, Alison Pouplin, Alice Del Vecchio, Suraj M S

Active learning (AL) is a sub-field of ML focused on the development of methods to iteratively and economically acquire data by strategically querying new data points that are the most useful for a particular task. Here, we introduce PyRelationAL, an open source library for AL research. We describe a modular toolkit based around a two step design methodology

Wasi Uddin Ahmad, Saikat Chakraborty, Baishakhi Ray, Kai-Wei Chang

Back-translation is widely known for its effectiveness in neural machine translation when there is little to no parallel data. In this approach, a source-to-target model is coupled with a target-to-source model trained in parallel. The target-to-source model generates noisy sources, while the source-to-target model is trained to reconstruct the targets and v

Manxi Lin, Zahra Bashir, Martin Grønnebæk Tolsgaard, Anders Nymark Christensen

Curvilinear structure segmentation is important in medical imaging, quantifying structures such as vessels, airways, neurons, or organ boundaries in 2D slices. Segmentation via pixel-wise classification often fails to capture the small and low-contrast curvilinear structures. Prior topological information is typically used to address this problem, often at a

Matthieu Romagny

We study algebraicity and smoothness of fixed point stacks for flat group schemes which have a finite composition series whose factors are either reductive or proper, flat, finitely presented, acting on algebraic stacks with affine, finitely presented diagonal. For this, we extend some theorems of [SGA3.2] on functors of homomorphisms Hom(G, H) and functors

Prisca Piccirilli, Sabine Schulte im Walde

Given a specific discourse, which discourse properties trigger the use of metaphorical language, rather than using literal alternatives? For example, what drives people to say "grasp the meaning" rather than "understand the meaning" within a specific context? Many NLP approaches to metaphorical language rely on cognitive and (psycho-)linguistic insights and

Arthur G. Suvorov, Hao-Jui Kuan, Kostas D. Kokkotas

GRB 211211A was accompanied by a kilonova, implying a merger origin for the event. A precursor flare, modulated by quasi-periodic oscillations at $\approx 22\, \mbox{Hz}$, was also seen $\approx 1\, \mbox{s}$ prior to the main emission. It is suggested here that the precursor resulted from the resonant shattering of one star's crust prior to coalescence. Sei

Cyrile Delestre, Abibatou Amar

Modern Natural Language Processing (NLP) models based on Transformer structures represent the state of the art in terms of performance on very diverse tasks. However, these models are complex and represent several hundred million parameters for the smallest of them. This may hinder their adoption at the industrial level, making it difficult to scale up to a

Pierre Lescanne

Holonomic equations are recursive equations which allow computing efficiently numbers of combinatoric objects. R{\'e}my showed that the holonomic equation associated with binary trees yields an efficient linear random generator of binary trees. I extend this paradigm to Motzkin trees and Schr{\"o}der trees and show that despite slight differences my algorith

Ning Gao, Jingyu Zhang, Ruijie Chen, Ngo Anh Vien

Grasping inhomogeneous objects in real-world applications remains a challenging task due to the unknown physical properties such as mass distribution and coefficient of friction. In this study, we propose a meta-learning algorithm called ConDex, which incorporates Conditional Neural Processes (CNP) with DexNet-2.0 to autonomously discern the underlying physi

Woo-Jeoung Nam, Seong-Whan Lee

With the remarkable success of deep neural networks, there is a growing interest in research aimed at providing clear interpretations of their decision-making processes. In this paper, we introduce Attribution Equilibrium, a novel method to decompose output predictions into fine-grained attributions, balancing positive and negative relevance for clearer visu

Petchiammal A, Briskline Kiruba S, D. Murugan, Pandarasamy A

One of the critical biotic stress factors paddy farmers face is diseases caused by bacteria, fungi, and other organisms. These diseases affect plants' health severely and lead to significant crop loss. Most of these diseases can be identified by regularly observing the leaves and stems under expert supervision. In a country with vast agricultural regions and

Lara Scavuzzo, Feng Yang Chen, Didier Chételat, Maxime Gasse

State-of-the-art Mixed Integer Linear Program (MILP) solvers combine systematic tree search with a plethora of hard-coded heuristics, such as the branching rule. The idea of learning branching rules from data has received increasing attention recently, and promising results have been obtained by learning fast approximations of the strong branching expert. In

Niki Kiriakidou, Christos Diou

Nowadays, in many scientific and industrial fields there is an increasing need for estimating treatment effects and answering causal questions. The key for addressing these problems is the wealth of observational data and the processes for leveraging this data. In this work, we propose a new model for predicting the potential outcomes and the propensity scor

Michal Hrbek

In a compactly generated triangulated category, we introduce a class of tilting objects satisfying certain purity condition. We call these the decent tilting objects and show that the tilting heart induced by any such object is equivalent to a category of contramodules over the endomorphism ring of the tilting object endowed with a natural linear topology. T

Marco Pleines, Matthias Pallasch, Frank Zimmer, Mike Preuss

At first sight it may seem straightforward to use recurrent layers in Deep Reinforcement Learning algorithms to enable agents to make use of memory in the setting of partially observable environments. Starting from widely used Proximal Policy Optimization (PPO), we highlight vital details that one must get right when adding recurrence to achieve a correct an

Rob Sumners, Cuong Chau

This volume contains a selection of papers presented at the 17th International Workshop on the ACL2 Theorem Prover and its Applications (ACL2 2022). The workshops are the premier technical forum for presenting research and experiences related to ACL2.

Jannik Castenow, Björn Feldkord, Till Knollmann, Manuel Malatyali

The $k$-Server Problem covers plenty of resource allocation scenarios, and several variations have been studied extensively for decades. We present a model generalizing the $k$-Server Problem by preferences of the requests, where the servers are not identical and requests can express which specific servers should serve them. In our model, requests can either

Sharut Gupta, Kartik Ahuja, Mohammad Havaei, Niladri Chatterjee

Federated learning aims to train predictive models for data that is distributed across clients, under the orchestration of a server. However, participating clients typically each hold data from a different distribution, whereby predictive models with strong in-distribution generalization can fail catastrophically on unseen domains. In this work, we argue tha

Jiangmeng Li, Wenyi Mo, Wenwen Qiang, Bing Su

Vision-language models are pre-trained by aligning image-text pairs in a common space to deal with open-set visual concepts. To boost the transferability of the pre-trained models, recent works adopt fixed or learnable prompts, i.e., classification weights are synthesized from natural language describing task-relevant categories, to reduce the gap between ta

Akiyoshi Sannai, Yasunari Hikima, Ken Kobayashi, Akinori Tanaka

In this paper, we propose a strategy to construct a multi-objective optimization algorithm from a single-objective optimization algorithm by using the B\'ezier simplex model. Also, we extend the stability of optimization algorithms in the sense of Probability Approximately Correct (PAC) learning and define the PAC stability. We prove that it leads to an uppe

Darren A. Thompson, Yakov I. Nesterets, Konstantin M. Pavlov, Timur E. Gureyev

A new method is developed for three-dimensional (3D) reconstruction of multi-material objects using propagation-based X-ray phase-contrast tomography (PB-CT) with phase retrieval via the contrast transfer function (CTF) formalism. The approach differs from conventional PB-CT algorithms that apply phase retrieval on individual two-dimensional (2D) projections

Linfeng Zhang, Runpei Dong, Hung-Shuo Tai, Kaisheng Ma

The remarkable breakthroughs in point cloud representation learning have boosted their usage in real-world applications such as self-driving cars and virtual reality. However, these applications usually have an urgent requirement for not only accurate but also efficient 3D object detection. Recently, knowledge distillation has been proposed as an effective m

Lijie Wang, Yaozong Shen, Shuyuan Peng, Shuai Zhang

While there is increasing concern about the interpretability of neural models, the evaluation of interpretability remains an open problem, due to the lack of proper evaluation datasets and metrics. In this paper, we present a novel benchmark to evaluate the interpretability of both neural models and saliency methods. This benchmark covers three representativ

Tobias Bernecker, Annette Peters, Christopher L. Schlett, Fabian Bamberg

Given the high incidence and effective treatment options for liver diseases, they are of great socioeconomic importance. One of the most common methods for analyzing CT and MRI images for diagnosis and follow-up treatment is liver segmentation. Recent advances in deep learning have demonstrated encouraging results for automatic liver segmentation. Despite th

Ilaria Balossino

A recently approved ten-year extension of the BESIII experiment (IHEP, Beijing) motivated an upgrade program for both the accelerator and the detector. In particular, the current inner drift chamber is suffering from aging and the proposal is to replace it with a detector based on the cylindrical GEM technology. The CGEM inner tracker (CGEM-IT) consists of t

Hongwei Gong, Haotian Liu, Bolong Jiao, Haoyi Zhang

The many-body physics in the dimensional crossover regime attracts much attention in cold atom experiments, but yet to explore systematically. One of the technical difficulties existed in the experiments is the lack of the experimental technique to quantitatively tune the atom occupation ratio of the different lattice bands. In this letter, we report such te

TaeHo Yoon, Ernest K. Ryu

Several new accelerated methods in minimax optimization and fixed-point iterations have recently been discovered, and, interestingly, they rely on a mechanism distinct from Nesterov's momentum-based acceleration. In this work, we show that these accelerated algorithms exhibit what we call the merging path (MP) property; the trajectories of these algorithms m

P. Chigansky, M. Kleptsyna

This paper addresses the problem of estimating the Hurst exponent of the fractional Brownian motion from continuous time noisy sample. Consistent estimation in the setup under consideration is possible only if either the length of the observation interval increases to infinity or intensity of the noise decreases to zero. The main result is a proof of the Loc

Giulia Murtas, Andrew Hillier, Ben Snow

Plasmoid-mediated fast magnetic reconnection plays a fundamental role in driving explosive dynamics and heating, but relatively little is known about how it develops in partially ionised plasmas (PIP) of the solar chromosphere. Partial ionisation might largely alter the dynamics of the coalescence instability, which promotes fast reconnection and forms a tur

Kai Wang, Bo Zhao, Xiangyu Peng, Zheng Zhu

Face recognition, as one of the most successful applications in artificial intelligence, has been widely used in security, administration, advertising, and healthcare. However, the privacy issues of public face datasets have attracted increasing attention in recent years. Previous works simply mask most areas of faces or synthesize samples using generative m

Xingyu Zhou, Gregory J. Herczeg, Yao Liu, Min Fang

The Serpens Molecular Cloud is one of the most active star-forming regions within 500 pc, with over one thousand of YSOs at different evolutionary stages. The ages of the member stars inform us about the star formation history of the cloud. In this paper, we develop a spectral energy distribution (SED) fitting method for nearby evolved (diskless) young stars

Yiming Chen, Yushen Fu, Mingyen Lee, Sumitha George

Compute-in-memory (CiM) is a promising approach to improving the computing speed and energy efficiency in dataintensive applications. Beyond existing CiM techniques of bitwise logic-in-memory operations and dot product operations, this paper extends the CiM paradigm with FAST, a new shift-based inmemory computation technique to handle high-concurrency operat

Nam H. Chu, Dinh Thai Hoang, Diep N. Nguyen, Khoa T. Phan

Creating and maintaining the Metaverse requires enormous resources that have never been seen before, especially computing resources for intensive data processing to support the Extended Reality, enormous storage resources, and massive networking resources for maintaining ultra high-speed and low-latency connections. Therefore, this work aims to propose a nov

Shijing Cheng, Wenting Zhou, Hongwei Yu

In a quantum sense, vacuum is not an empty void but full of virtual particles (fields). It may have long-range properties, be altered, and even undergo phase transitions. It is suggested that long-range properties of a quantum vacuum may be probed by distributing matter over a large spatial volume. Here, we study a simplest example of such, i.e., two uniform

Zhao Liu, Biao Wang, Dewei Chu, Claudio Cazorla

Finding cost-effective and efficient photocatalytic materials able to catalyse the water splitting reaction under visible light is one of the greatest challenges in current environmental material science. Despite that many photocatalysts are already known in the context of green hydrogen production, strategies to systematically and rationally modify their op

Blerina Sinaimeri, Laura Urbini, Marie-France Sagot, Catherine Matias

Phylogenetic tree reconciliation is employed for the examination of coevolution between host and symbiont species. An important concern is the requirement for dependable cost values when selecting event-based parsimonious reconciliation. Although certain approaches deduce event probabilities unique to each pair of host and symbiont trees, which can subsequen

Guruprasad Deshpande, Mangesh Goswami, Jayesh Kolhe, Vishal Khandagale

Soil moisture (SM) is referred to as a finite amount of water molecules within the pore spaces and it is a crucial parameter of Hydro-Meteorological processes. The behaviour of soil moisture water changes spatially and temporally in response to topography, soil characteristics, and climate[1]. Soil moisture is overseen by various hydro-meteorological factors

Jinwoo Bae, Sungho Moon, Sunghoon Im

Self-supervised monocular depth estimation has been widely studied recently. Most of the work has focused on improving performance on benchmark datasets, such as KITTI, but has offered a few experiments on generalization performance. In this paper, we investigate the backbone networks (e.g. CNNs, Transformers, and CNN-Transformer hybrid models) toward the ge

Tanvi Mehta, Ganesh Deshmukh

Sentiment Analysis is currently a vital area of research. With the advancement in the use of the internet, the creation of social media, websites, blogs, opinions, ratings, etc. has increased rapidly. People express their feedback and emotions on social media posts in the form of likes, dislikes, comments, etc. The rapid growth in the volume of viewer-genera

A. D. Patel, Zubin Shaikh, M. Sharma, Santosh P. Pandya

A Multi-dipole line cusp configured Plasma Device (MPD) having six electromagnets with embedded Vacoflux-50 as a core material and a hot filament-based cathode for Argon plasma production has been characterized by changing the pole magnetic field values. For the next step ahead, a new tungsten ionizer plasma source for contact ionization cesium plasma has be

Abhik Bhattacharjee, Tahmid Hasan, Wasi Uddin Ahmad, Rifat Shahriyar

This work presents BanglaNLG, a comprehensive benchmark for evaluating natural language generation (NLG) models in Bangla, a widely spoken yet low-resource language. We aggregate six challenging conditional text generation tasks under the BanglaNLG benchmark, introducing a new dataset on dialogue generation in the process. Furthermore, using a clean corpus o

Hai-Long Shi, Shu Ding, Qing-Kun Wan, Xiao-Hui Wang

We investigate the connection between quantum resources and extractable work in quantum batteries. We demonstrate that quantum coherence in the battery or the battery-charger entanglement is a necessary resource for generating nonzero extractable work during the charging process. At the end of the charging process, we also establish a tight link of coherence