May 2024 arXiv papers — page 51

Tianyi Chen, Jianfu Zhang, Yan Hong, Yiyi Zhang

Image inpainting, the task of reconstructing missing segments in corrupted images using available data, faces challenges in ensuring consistency and fidelity, especially under information-scarce conditions. Traditional evaluation methods, heavily dependent on the existence of unmasked reference images, inherently favor certain inpainting outcomes, introducin

Runqi Lin, Chaojian Yu, Bo Han, Hang Su

Catastrophic overfitting (CO) presents a significant challenge in single-step adversarial training (AT), manifesting as highly distorted deep neural networks (DNNs) that are vulnerable to multi-step adversarial attacks. However, the underlying factors that lead to the distortion of decision boundaries remain unclear. In this work, we delve into the specific

Mikhail S. Podoshvedov, Sergey A. Podoshvedov

We study sensitivity of phase estimation of Mach-Zehnder (MZ) interferometer with original two-mode squeezed vacuum (TMSV) state. At the initial stage, the TMSV state is converted into two single-mode squeezed vacuum (SMSV) states, from each of which photons are subtracted by measurement by photon-number resolving (PNR) detector in auxiliary modes. New measu

Shelly Golan, Roy Ganz, Michael Elad

The recently introduced Consistency models pose an efficient alternative to diffusion algorithms, enabling rapid and good quality image synthesis. These methods overcome the slowness of diffusion models by directly mapping noise to data, while maintaining a (relatively) simpler training. Consistency models enable a fast one- or few-step generation, but they

Javier Viaña

This paper introduces the front-propagation algorithm, a novel eXplainable AI (XAI) technique designed to elucidate the decision-making logic of deep neural networks. Unlike other popular explainability algorithms such as Integrated Gradients or Shapley Values, the proposed algorithm is able to extract an accurate and consistent linear function explanation o

Hong Liu, Xiuxiu Qiu, Yiming Shi, Miao Xu

Unsupervised fault detection in multivariate time series plays a vital role in ensuring the stable operation of complex systems. Traditional methods often assume that normal data follow a single Gaussian distribution and identify anomalies as deviations from this distribution. {\color{black} However, this simplified assumption fails to capture the diversity

Wanli Ni, Ailing Zheng, Wen Wang, Dusit Niyato

Although reconfigurable intelligent surfaces (RISs) have demonstrated the potential to boost network capacity and expand coverage by adjusting their electromagnetic properties, existing RIS architectures have certain limitations, such as double-fading attenuation and restricted half-space coverage. In this article, we delve into the progressive development f

Si Xu, Zixiao Huang, Yan Zeng, Shengen Yan

Training large-scale models relies on a vast number of computing resources. For example, training the GPT-4 model (1.8 trillion parameters) requires 25000 A100 GPUs . It is a challenge to build a large-scale cluster with one type of GPU-accelerator. Using multiple types of GPU-accelerators to construct a large-scale cluster is an effective way to solve the p

Chinedu Eleh, Masuzyo Mwanza, Ekene Aguegboh, Hans-Werner van Wyk

The Adam optimization method has achieved remarkable success in addressing contemporary challenges in stochastic optimization. This method falls within the realm of adaptive sub-gradient techniques, yet the underlying geometric principles guiding its performance have remained shrouded in mystery, and have long confounded researchers. In this paper, we introd

Philipp Schmidt, Florian Marquardt, Naeimeh Mohseni

Deep neural networks have demonstrated remarkable efficacy in extracting meaningful representations from complex datasets. This has propelled representation learning as a compelling area of research across diverse fields. One interesting open question is how beneficial representation learning can be for quantum many-body physics, with its notouriosly high-di

Zeling Shao, Xiaoxiang Yu, Zhiguo Li

In this paper, the dispersability of the Cartesian graph bundle over two cycles is completely solved. We show the Cartesian graph bundle $G$ over two cycles is dispersable if $G$ is bipartite; otherwise, $G$ is nearly dispersable.

Zhenkun Li, Fan Ye

This paper studies the existence of $2$-torsion in instanton Floer homology with $\mathbb{Z}$ coefficients for closed $3$-manifolds and singular knots. First, we show that the non-existence of $2$-torsion in the framed instanton Floer homology $I^\sharp(S_n^3(K);\mathbb{Z})$ of any nonzero integral $n$-surgery along a knot $K$ in $S^3$ would imply that $K$ i

Meng-Kiat Chuah, Rita Fioresi

Geometric quantization transforms a symplectic manifold with Lie group action to a unitary representation. In this article, we extend geometric quantization to the super setting. We consider real forms of contragredient Lie supergroups with compact Cartan subgroups, and study their actions on some pseudo-K\"ahler supermanifolds. We construct their unitary re

Yash Patel, Sahana Rayan, Ambuj Tewari

End-to-end engineering design pipelines, in which designs are evaluated using concurrently defined optimal controllers, are becoming increasingly common in practice. To discover designs that perform well even under the misspecification of system dynamics, such end-to-end pipelines have now begun evaluating designs with a robust control objective in place of

Siran Li, Xiangxiang Su

A fundamental result in global analysis and nonlinear elasticity asserts that given a solution $\mathfrak{S}$ to the Gauss--Codazzi--Ricci equations over a simply-connected closed manifold $(\mathcal{M}^n,g)$, one may find an isometric immersion $\iota$ of $(\mathcal{M}^n,g)$ into the Euclidean space $\mathbb{R}^{n+k}$ whose extrinsic geometry coincides with

Junlin Song, Yuzhuo Chen, Yuan Yao, Zetong Chen

Autism Spectrum Disorder is a condition characterized by a typical brain development leading to impairments in social skills, communication abilities, repetitive behaviors, and sensory processing. There have been many studies combining brain MRI images with machine learning algorithms to achieve objective diagnosis of autism, but the correlation between whit

Minghao Chen, Yihang Li, Yanting Yang, Shiyu Yu

Large Language Models (LLM) based agents have shown promise in autonomously completing tasks across various domains, e.g., robotics, games, and web navigation. However, these agents typically require elaborate design and expert prompts to solve tasks in specific domains, which limits their adaptability. We introduce AutoManual, a framework enabling LLM agent

Eduardo Ochoa Rivera, Yash Patel, Ambuj Tewari

Distribution-free uncertainty estimation for ensemble methods is increasingly desirable due to the widening deployment of multi-modal black-box predictive models. Conformal prediction is one approach that avoids such distributional assumptions. Methods for conformal aggregation have in turn been proposed for ensembled prediction, where the prediction regions

Jing Li, Kaiqiang Wen, Ke Xiao, Xiaoming Chen

Due to the potential application of regulating droplet shape by external fields in microfluidic technology and micro devices, it becomes increasingly important to understand the shape formation of a droplet in the presence of an electric field. How to understand and determine such a deformable boundary shape at equilibrium has been a long-term physical and m

D. Tsiklauri

We use particle-in-cell, fully electromagnetic, plasma kinetic simulation to study the effect of external magnetic field on electron scale Kelvin-Helmholtz instability (ESKHI). The results are applicable to collisionless plasmas when e.g. solar wind interacts with planetary magnetospheres or magnetic field is generated in AGN jets. We find that as in the cas

Ethan Akin, Marian Mrozek, Mateusz Przybylski, Jim Wiseman

We give a complete invariant for shift equivalence for Boolean matrices (equivalently finite relations), in terms of the period, the induced partial order on recurrent components, and the cohomology class of the relation on those components.

Gongming Yu, Rabia Hameed, Liyuan Hu, Qiang Hu

We have investigated the single and double diffractive production of dileptons and photons in ultra-peripheral collisions at the Large Hadron Collider (LHC). Utilizing advanced theoretical models that integrate quantum electrodynamics (QED) and Quantum Chromodynamics (QCD) frameworks, we analyze the differential cross sections of these processes, with partic

Chenqi Lin, Tianshi Xu, Zebin Yang, Runsheng Wang

With the fast evolution of large language models (LLMs), privacy concerns with user queries arise as they may contain sensitive information. Private inference based on homomorphic encryption (HE) has been proposed to protect user query privacy. However, a private embedding table query has to be formulated as a HE-based matrix-vector multiplication problem an

Run He, Kai Tong, Di Fang, Han Sun

In this paper, we introduce analytic federated learning (AFL), a new training paradigm that brings analytical (i.e., closed-form) solutions to the federated learning (FL) with pre-trained models. Our AFL draws inspiration from analytic learning -- a gradient-free technique that trains neural networks with analytical solutions in one epoch. In the local clien

Aliaksei Kachanovich, Ivan Nišandžić

We present new calculations of the leading-order one-loop amplitude and the decay rate for $H\to \nu \bar{\nu} \gamma$ within the standard model, employing reduction to a set of basic Passarino-Veltman functions resulting in more concise expressions compared to prior works. Our result for the total decay rate is $\Gamma(H\to \nu\bar{\nu} \gamma)=1.33\,\text{

Stephan Stadler

We characterize higher rank model geometries -- Riemannian symmetric spaces, Euclidean buildings and products -- among Hadamard spaces by using antipodal sets at infinity.

Philippe Weier, Alexander Rath, Élie Michel, Iliyan Georgiev

Neural representations have shown spectacular ability to compress complex signals in a fraction of the raw data size. In 3D computer graphics, the bulk of a scene's memory usage is due to polygons and textures, making them ideal candidates for neural compression. Here, the main challenge lies in finding good trade-offs between efficient compression and cheap

Seamus Somerstep, Felipe Maia Polo, Moulinath Banerjee, Ya'acov Ritov

Modern large language model (LLM) alignment techniques rely on human feedback, but it is unclear whether these techniques fundamentally limit the capabilities of aligned LLMs. In particular, it is unknown if it is possible to align (stronger) LLMs with superhuman capabilities with (weaker) human feedback without degrading their capabilities. This is an insta

Yuzhuo Chen, Zetong Chen, Yuanyuan Liu

Abnormalities in retinal fundus images may indicate certain pathologies such as diabetic retinopathy, hypertension, stroke, glaucoma, retinal macular edema, venous occlusion, and atherosclerosis, making the study and analysis of retinal images of great significance. In conventional medicine, the diagnosis of retina-related diseases relies on a physician's su

Shiyu Xia, Junyu Xiong, Haoyu Dong, Jianbo Zhao

This paper explores capabilities of Vision Language Models on spreadsheet comprehension. We propose three self-supervised challenges with corresponding evaluation metrics to comprehensively evaluate VLMs on Optical Character Recognition (OCR), spatial perception, and visual format recognition. Additionally, we utilize the spreadsheet table detection task to

Yongxin Guo, Lin Wang, Xiaoying Tang, Tao Lin

Federated Learning (FL) is a privacy-preserving distributed machine learning paradigm. Nonetheless, the substantial distribution shifts among clients pose a considerable challenge to the performance of current FL algorithms. To mitigate this challenge, various methods have been proposed to enhance the FL training process. This paper endeavors to tackle the i

Shuaibin Gao, Qian Guo, Zhuoqi Liu, Chenggui Yuan

This paper focuses on the numerical scheme for delay-type stochastic McKean-Vlasov equations (DSMVEs) driven by fractional Brownian motion with Hurst parameter $H\in (0,1/2)\cup (1/2,1)$. The existence and uniqueness of the solutions to such DSMVEs whose drift coefficients contain polynomial delay terms are proved by exploting the Banach fixed point theorem.

Gábor Hegedüs

Let $\mbox{$\cal V$} \subseteq {\mathbb F}^n$ be a finite set of points in an affine space. A finite set of affine hyperplanes $\{H_1, \ldots ,H_m\}$ is said to be an almost cover of $\mbox{$\cal V$}$ and $\mathbf{v}$, if their union $\cup_{j=1}^m H_j$ contains $\mbox{$\cal V$}\setminus \{\mathbf{v}\}$ but does not contain $\mathbf{v}$. We give here a lower

Sumner B. Harris, Rama Vasudevan, Yongtao Liu

The fusion of experimental automation and machine learning has catalyzed a new era in materials research, prominently featuring Gaussian Process Bayesian Optimization (GPBO) driven autonomous experiments navigating complex experimental conditions for accelerated scientific discovery. In traditional GPBO-driven experiments, a predefined scalarizer function is

Chak Tou Leong, Yi Cheng, Kaishuai Xu, Jian Wang

The existing safety alignment of Large Language Models (LLMs) is found fragile and could be easily attacked through different strategies, such as through fine-tuning on a few harmful examples or manipulating the prefix of the generation results. However, the attack mechanisms of these strategies are still underexplored. In this paper, we ask the following qu

Antonin Sulc, Alex Bien, Annika Eichler, Daniel Ratner

Electronic logbooks contain valuable information about activities and events concerning their associated particle accelerator facilities. However, the highly technical nature of logbook entries can hinder their usability and automation. As natural language processing (NLP) continues advancing, it offers opportunities to address various challenges that logboo

Cezhong Tong, Xin He, Zicong Yang

The Volterra-type integral operator plays an essential role in modern complex analysis and operator theory. Recently, Chalmoukis \cite{Cn} introduced a generalized integral operator, say $I_{g,a}$, defined by $$I_{g,a}f=I^n(a_0f^{(n-1)}g'+a_1f^{(n-2)}g''+\cdots+a_{n-1}fg^{(n)}),$$ where $g\in H(\mathbb{D})$ and $a=(a_0,a_1,\cdots,a_{n-1})\in \mathbb{C}^n$. $

Gennady Khalimov, Yevgen Kotukh, Maksym Kolisnyk, Svitlana Khalimova

The paper explores a novel cryptosystem for digital signatures based on linear equa-tions for logarithmic signatures. A logarithmic signature serves as a fundamental cryptographic primitive, characterized by properties such as nonlinearity, non-commutability, unidirectionality, and key-dependent factorability. The proposed cryptosystem ensures the secrecy of

Qian Wang, Chen Li, Yuchen Luo, Hefei Ling

As a defense strategy against adversarial attacks, adversarial detection aims to identify and filter out adversarial data from the data flow based on discrepancies in distribution and noise patterns between natural and adversarial data. Although previous detection methods achieve high performance in detecting gradient-based adversarial attacks, new attacks b

Feng Xie, Zheng Li, Peng Wu, Yan Zeng

Discovering causal relationships from observational data, particularly in the presence of latent variables, poses a challenging problem. While current local structure learning methods have proven effective and efficient when the focus lies solely on the local relationships of a target variable, they operate under the assumption of causal sufficiency. This as

Zixu Wang, Bingbing Xu, Yige Yuan, Huawei Shen

Graph contrastive learning (GCL), standing as the dominant paradigm in the realm of graph pre-training, has yielded considerable progress. Nonetheless, its capacity for out-of-distribution (OOD) generalization has been relatively underexplored. In this work, we point out that the traditional optimization of InfoNCE in GCL restricts the cross-domain pairs onl

Somnath Pradhan, Serdar Yuksel

For optimal control of diffusions under several criteria, due to computational or analytical reasons, many studies have a apriori assumed control policies to be Lipschitz or smooth, often with no rigorous analysis on whether this restriction entails loss. While optimality of Markov/stationary Markov policies for expected finite horizon/infinite horizon (disc

Yu Cheng, Jiadu Lin, Jie Sheng, Tsutomu T. Yanagida

The experimental verification of the Newton law of gravity at small scales has been a longstanding challenge. Recently, torsion balance experiments have successfully measured gravitational force at the millimeter scale. However, testing gravity force on quantum mechanical wave function at small scales remains difficult. In this paper, we propose a novel expe

Rizwan Ahmed, Hazrat Ali, Aamir Shehzad, S K Singh

We propose a theoretical scheme for the generation of nonreciprocal multipartite entanglement in a two-mode cavity magnomechanical system, consisting of two cross-microwave (MW) cavities having a yttrium iron garnet (YIG) sphere, which is coupled through magnetic dipole interaction. Our results show that the self-Kerr effect of magnon can significantly enhan

Yuzhuo Chen, Zetong Chen, Yunuo An, Chenyang Lu

The precise categorization of white blood cell (WBC) is crucial for diagnosing blood-related disorders. However, manual analysis in clinical settings is time-consuming, labor-intensive, and prone to errors. Numerous studies have employed machine learning and deep learning techniques to achieve objective WBC classification, yet these studies have not fully ut

Qilong Zhao, Shiyu Wang, Zeeshan Memon, Yang Qiao

Controllable data generation aims to synthesize data by specifying values for target concepts. Achieving this reliably requires modeling the underlying generative factors and their relationships. In real-world scenarios, these factors exhibit both causal and correlational dependencies, yet most existing methods model only part of this structure. We propose t

Alexander Tyurin, Peter Richtárik

We consider the decentralized stochastic asynchronous optimization setup, where many workers asynchronously calculate stochastic gradients and asynchronously communicate with each other using edges in a multigraph. For both homogeneous and heterogeneous setups, we prove new time complexity lower bounds under the assumption that computation and communication

Peter Lundgaard, Andreas Bøgh Poulsen

In [4] Sturmfels linked the Hilbert Nullstellensatz to Gr\"obner bases through final polynomials. In (loc. cit.) it was claimed that final polynomials always appear in a lexicographic Gr\"obner basis of a certain ideal. In this paper, we give a counterexample to this claim. We also show how the introduction of an extra variable restores the claim in a deform

Ryan Benkert, Mohit Prabhushankar, Ghassan AlRegib

In this paper, we discuss feature engineering for single-pass uncertainty estimation. For accurate uncertainty estimates, neural networks must extract differences in the feature space that quantify uncertainty. This could be achieved by current single-pass approaches that maintain feature distances between data points as they traverse the network. While init

Christopher-Lloyd Simon, Ben Stucky

A multiloop $\gamma\colon \sqcup_1^s \mathbb{S}^1 \looparrowright \mathbb{F}$ is a generic immersion of a finite union of circles into an oriented surface, considered up to homeomorphisms. A pinning set is a set of points $P\subset \mathbb{F}\setminus \operatorname{im}(\gamma)$, such that in the punctured surface $\mathbb{F} \setminus P$, the immersion $\gam

Marten de Bruin, Konstantinos Mersinas

Cyber security incidents are increasing and humans play an important role in reducing their likelihood and impact. We identify a skewed focus towards technical aspects of cyber security in the literature, whereas factors influencing the secure behaviour of individuals require additional research. These factors span across both the individual level and the co

Shuaixin Liu, Kunqian Li, Yilin Ding, Qi Qi

Underwater Image Enhancement (UIE) aims to improve the visual quality from a low-quality input. Unlike other image enhancement tasks, underwater images suffer from the unavailability of real reference images. Although existing works exploit synthetic images and manually select well-enhanced images as reference images to train enhancement networks, their uppe

Gabriel Moreira, Manuel Marques, João Paulo Costeira, Alexander Hauptmann

Learning image representations that capture rich semantic relationships remains a significant challenge. Existing approaches are either contrastive, lacking robust theoretical guarantees, or struggle to effectively represent the partial orders inherent to structured visual-semantic data. In this paper, we introduce a nuclear norm-based loss function, grounde

Ali Zamani

Several numerical radius inequalities in the framework of $C^*$-algebras are proved in this paper. These results, which are based on an extension of Buzano inequality for elements in a pre-Hilbert $C^*$-module, generalize earlier numerical radius inequalities.

Mohammad Mohammadi Sabet

The q-deformed statistical mechanics for fermions has been used to investigate the Thomas-Fermi screening length at finite temperature. Considering linear response, the calculations have been made at weakly nondegenerate regime. The results show that q-deformation has significance effects on screening length at higher temperatures. It is also shown that the

Jingwei Li, Jing Dong, Tianxing He, Jingzhao Zhang

Given the rising popularity of AI-generated art and the associated copyright concerns, identifying whether an artwork was used to train a diffusion model is an important research topic. The work approaches this problem from the membership inference attack (MIA) perspective. We first identify the limitation of applying existing MIA methods for proprietary dif

Muhammad Omar Nadeem

We study the contribution of Scalar Leptoquark (SLQ) loops to the two photon scattering cross section at high energies. The leading order helicity amplitudes are discussed at coupling O(e4) for Standard Model (SM) fermions, weak bosons and SLQs. Helicity amplitudes for the different types of loops are calculated for photon center of mass energies being much

Jianyong Wei, Yumeng Liu, Yizhuo Wang, Kai Li

High-gain photodetectors based on two-dimensional (2D) semiconductors, in particular those in photoconductive mode, have been extensively investigated in the past decade. However, the classical photoconductive theory was derived on two misplaced assumptions. In this work, we established an explicit analytical device model for Schottky contact MoS2 phototrans

Aaron Brown, Alex Eskin, Simion Filip, Federico Rodriguez Hertz

We revisit the theory of normal forms for non-uniformly contracting dynamics. We collect a number of lemmas and reformulations of the standard theory that will be used in other projects.

Haiyang Shi

The increasing frequency of global climate extremes has significantly impacted the terrestrial carbon cycle. Extreme weather events such as heatwaves, droughts, and extreme precipitation pose serious threats to ecosystem carbon sequestration. This study investigated the impacts of these extreme events on terrestrial carbon sequestration using data from weath

Minghao Xu, Yunteng Geng, Yihang Zhang, Ling Yang

Glycans are basic biomolecules and perform essential functions within living organisms. The rapid increase of functional glycan data provides a good opportunity for machine learning solutions to glycan understanding. However, there still lacks a standard machine learning benchmark for glycan property and function prediction. In this work, we fill this blank

Zhizheng Wang, Qiao Jin, Chih-Hsuan Wei, Shubo Tian

Gene set knowledge discovery is essential for advancing human functional genomics. Recent studies have shown promising performance by harnessing the power of Large Language Models (LLMs) on this task. Nonetheless, their results are subject to several limitations common in LLMs such as hallucinations. In response, we present GeneAgent, a first-of-its-kind lan

Phong Tran, Egor Zakharov, Long-Nhat Ho, Liwen Hu

We introduce VOODOO XP: a 3D-aware one-shot head reenactment method that can generate highly expressive facial expressions from any input driver video and a single 2D portrait. Our solution is real-time, view-consistent, and can be instantly used without calibration or fine-tuning. We demonstrate our solution on a monocular video setting and an end-to-end VR

Nanxu Gong, Chandan K. Reddy, Wangyang Ying, Haifeng Chen

Feature transformation aims to reconstruct the feature space of raw features to enhance the performance of downstream models. However, the exponential growth in the combinations of features and operations poses a challenge, making it difficult for existing methods to efficiently explore a wide space. Additionally, their optimization is solely driven by the a

Aaron Brown, Federico Rodriguez Hertz, Zhiren Wang

We study actions by lattices in higher-rank (semi)simple Lie groups on compact manifolds. By classifying certain measures invariant under a related higher-rank abelian action (the diagonal action on the suspension space) we deduce a number of new rigidity results related to standard projective actions (i.e. boundary actions) by such groups. Specifically, in

Jiayu Liu, Tingting Luo, Cairong Chen

In this paper, we reconsider two new iterative methods for solving absolute value equations (AVE), which is proposed by Ali and Pan (Jpn. J. Ind. Appl. Math. 40: 303--314, 2023). Convergence results of the two iterative schemes and new sufficient conditions for the unique solvability of AVE are presented. In addition, for a special case, the optimal iteratio

Lan Wu, Xuebin Wang, Ruijuan Chu, Guangyi Liu

Accurate multi-step flight trajectory prediction plays an important role in Air Traffic Control, which can ensure the safety of air transportation. Two main issues limit the flight trajectory prediction performance of existing works. The first issue is the negative impact on prediction accuracy caused by the significant differences in data range. The second

Shudan Xue, Qingying Deng

Twisted knot theory, introduced by M.O. Bourgoin, is a generalization of virtual knot theory. It naturally yields the notion of a twisted braid, which is closely related to the notion of a virtual braid due to Kauffman. In this paper, we first prove that any twisted link can be described as the closure of a twisted braid, which is unique up to certain basic

Arijit Mukherjee

For a partition $(n_1,\ldots,n_r)$ of a positive integer $n$, consider the associated multiprojective space $\mathbb{P}^{n_1}\times\cdots\times\mathbb{P}^{n_r}$. That multiprojective spaces attached to distinct partitions of $n$ are pairwise non-isomorphic is known, having been established by algebro-geometric methods. In this paper we give a new, representa

Fuheng Zhou, Dikai Wei, Ye Fan, Yulong Huang

Although deep learning based models for underwater image enhancement have achieved good performance, they face limitations in both lightweight and effectiveness, which prevents their deployment and application on resource-constrained platforms. Moreover, most existing deep learning based models use data compression to get high-level semantic information in l

Makgotso Jacqueline Maotwana

Poor roads are a major issue for cars, drivers, and pedestrians since they are a major cause of vehicle damage and can occasionally be quite dangerous for both groups of people (pedestrians and drivers), this makes road surface condition monitoring systems essential for traffic safety, reducing accident rates ad also protecting vehicles from getting damaged.

Théo Vincent, Fabian Wahren, Jan Peters, Boris Belousov

Deep Reinforcement Learning (RL) is well known for being highly sensitive to hyperparameters, requiring practitioners substantial efforts to optimize them for the problem at hand. This also limits the applicability of RL in real-world scenarios. In recent years, the field of automated Reinforcement Learning (AutoRL) has grown in popularity by trying to addre

Chun-Mao Lai, Hsiang-Chun Wang, Ping-Chun Hsieh, Yu-Chiang Frank Wang

Imitation learning aims to learn a policy from observing expert demonstrations without access to reward signals from environments. Generative adversarial imitation learning (GAIL) formulates imitation learning as adversarial learning, employing a generator policy learning to imitate expert behaviors and discriminator learning to distinguish the expert demons

Jieh-Sheng Lee

In this research, patent prosecution is conceptualized as a system of reinforcement learning from human feedback. The objective of the system is to increase the likelihood for a language model to generate patent claims that have a higher chance of being granted. To showcase the controllability of the language model, the system learns from granted patents and

Sammy Kemboi Chepkilot

In Kenya, interest payments on external debt have been increasing from 2010 to 2015, while GDP growth experienced a slight decline over the same period. Policymakers are concerned that the rapid increase in external debt in developing countries such as Kenya has the potential to erode the country's sovereign rating, particularly if it is not supported by pro

Roberto Vila, Eduardo Nakano, Helton Saulo

In this paper, we propose and investigate closed-form point estimators for a weighted exponential family. We also develop a bias-reduced version of these proposed closed-form estimators through bootstrap methods. Estimators are assessed using a Monte Carlo simulation, revealing favorable results for the proposed bootstrap bias-reduced estimators.

Junhao Yu, Jiarun Wei

In this project, we attempt to optimize a landing trajectory of a rocket. The goal is to minimize the total fuel consumption during the landing process using different techniques. Once the optimal and feasible trajectory is generated using batch approach, we attempt to follow the path using a Model Predictive Control (MPC) based algorithm, called Trajectory

Swapnil Kumar Singh

Quantum Chromodynamics (QCD) is the fundamental theory describing the strong nuclear force and the interactions among quarks and gluons. Topological stars, characterized by extreme density conditions, offer a unique environment where QCD phenomena play a crucial role due to the confinement of fundamental particles. Understanding these phenomena is essential

Ding Zhang, Gururaj V. Naik

Light-matter interaction in quantum materials presents a new paradigm as light can tip the balance between many competing quantum many-body phases to result in new phenomena. Describing the optical response of such materials requires complex models. Here, we develop a non-local model to describe the optical response of a quantum material, 1T-TaS$_2$. 1T-TaS$

Yongkang Yan, Peng Zhang, Qingzhong Liu, Zhi Chang

This study presents the detection of a high-frequency quasi-periodic oscillation (QPO) in the Seyfert galaxy NGC 1365 based on observational data obtained by \emph{XMM-Newton} in January 2004. Utilizing the weighted wavelet Z-transform (WWZ) and Lomb-Scargle periodogram (LSP) methods, a QPO signal is identified at a frequency of $2.19 \times 10^{-4}\ {\rm Hz

Juan Casado-Díaz, Nourelhouda Khedhiri, Mohamed Lazhar Tayeb

We consider a Maxwell system on $\mathbb{R}^3$ with periodic and highly oscillating coefficients. It is known that the solutions converge in the weak-$\ast$ topology of $L^\infty(0,T;\,L^2(\mathbb{R}^3))$ to the solution of a similar problem with constant coefficients given as the $H$-limits of the electric permittivity and the magnetic permeability respecti

Yanfei Dong, Mohammed Haroon Dupty, Lambert Deng, Zhuanghua Liu

Graph Neural Networks often struggle with long-range information propagation and in the presence of heterophilous neighborhoods. We address both challenges with a unified framework that incorporates a clustering inductive bias into the message passing mechanism, using additional cluster-nodes. Central to our approach is the formulation of an optimal transpor

Harry Zhang

Model-based Reinforcement Learning (MBRL) has shown many desirable properties for intelligent control tasks. However, satisfying safety and stability constraints during training and rollout remains an open question. We propose a new Model-based RL framework to enable efficient policy learning with unknown dynamics based on learning model predictive control (

Masanobu Horie, Naoto Mitsume

Utilizing machine learning to address partial differential equations (PDEs) presents significant challenges due to the diversity of spatial domains and their corresponding state configurations, which complicates the task of encompassing all potential scenarios through data-driven methodologies alone. Moreover, there are legitimate concerns regarding the gene

Amin A. Nizami, Ankit W. Shrestha

The complexity of quantum states under dynamical evolution can be investigated by studying the spread with time of the state over a pre-defined basis. It is known that this complexity is minimised by choosing the Krylov basis, thus defining the spread complexity. We study the dynamics of spread complexity for quantum maps using the Arnoldi iterative procedur

Chunlin Qiu, Ang Li, Yiheng Duan, Shenyi Zhang

Transfer-based attacks craft adversarial examples on white-box surrogate models and directly deploy them against black-box target models, offering model-agnostic and query-free threat scenarios. While flatness-enhanced methods have recently emerged to improve transferability by enhancing the loss surface flatness of adversarial examples, their divergent flat

Prakash Sarkar, Prita Pant, Hemant Nanavati

We identify for visco-elasto-plastic (VEP) glassy polymers, physical phenomena during Berkovich nanoindentation, a locally imposed deformation. Live visuals via in situ nanoindentation indicate mainly sink-in during loading, with pile-up after unloading. Scanning Probe Microscopy (SPM) indicates significant volume conserving upflow below the tip, for these h

Elisenda Feliu, Nidhi Kaihnsa

Phosphorylation networks, representing the mechanisms by which proteins are phosphorylated at one or multiple sites, are ubiquitous in cell signalling and display rich dynamics such as unlimited multistability. Dual-site phosphorylation networks are known to exhibit oscillations in the form of periodic trajectories, when phosphorylation and dephosphorylation

Yun Zhu, Jia-Chen Gu, Caitlin Sikora, Ho Ko

Large language models (LLMs) augmented with retrieval exhibit robust performance and extensive versatility by incorporating external contexts. However, the input length grows linearly in the number of retrieved documents, causing a dramatic increase in latency. In this paper, we propose a novel paradigm named Sparse RAG, which seeks to cut computation costs

Thanh-Dung Le, Clara Macabiau, Kévin Albert, Symeon Chatzinotas

This study delves into the effectiveness of various learning methods in improving Transformer models, focusing particularly on the Gated Residual Network Transformer (GRN-Transformer) in the context of pediatric intensive care units (PICU) with limited data availability. Our findings indicate that Transformers trained via supervised learning are less effecti

Łukasz Kamiński, Sławomir Lasota

We investigate Petri nets with data, an extension of plain Petri nets where tokens carry values from an infinite data domain, and executability of transitions is conditioned by equalities between data values. We provide a decision procedure for the bi-reachability problem: given a Petri net and its two configurations, we ask if each of the configurations is

Kamel Hamdache, Djamila Hamroun, Basma Jaffal-Mourtada

In this work we prove the existence of time-periodic solutions to a model describing a ferrofluid flow heated from below. Navier-Stokes equations satisfied by the fluid velocity are coupled to the temperature equation and the magnetostatic equation satisfied by the magnetic potential. The magnetization is assumed to be parallel to the magnetic field and is g

Davide Dardari

To meet the stringent requirements of next-generation wireless networks, multiple-input multiple-output (MIMO) technology is expected to become massive and pervasive. Unfortunately, this could pose scalability issues in terms of complexity, power consumption, cost, and processing latency. Therefore, novel technologies and design approaches, such as the recen

Shutong Ding, Ke Hu, Zhenhao Zhang, Kan Ren

Diffusion models have garnered widespread attention in Reinforcement Learning (RL) for their powerful expressiveness and multimodality. It has been verified that utilizing diffusion policies can significantly improve the performance of RL algorithms in continuous control tasks by overcoming the limitations of unimodal policies, such as Gaussian policies, and

Cairong Chen, Xuehua Li, Ren-Cang Li

An underdetermined generalized absolute value equation (GAVE) may have no solution, one solution, finitely many or infinitely many solutions. This paper is concerned with sufficient conditions that guarantee the existence of solutions to an underdetermined GAVE. Particularly, sufficient conditions are established for an underdetermined GAVE to have infinitel

A. A. Herasymchuk, S. G. Sharapov, V. P. Gusynin

Employing the low-energy effective theory alongside a combination of analytical and numerical techniques, we explore the Landau level collapse phenomenon, uncovering previously undisclosed features. We consider both finite-width graphene ribbons and semi-infinite geometries subjected to a perpendicular magnetic field and an in-plane electric field, applied p

Lorenzo Pica

The LHCb experiment is starting to take data in Run 3 with a new DAQ system, capable of performing complete event reconstruction at the full LHC collision rate. One novel opportunity offered by this system is triggering on long-lived particles (LLPs) at the very first stage of the trigger. This could potentially increase trigger efficiency for LLPs, typicall

André M. Sonnet, Epifanio G. Virga

Minimal surfaces are ubiquitous in nature. Here they are considered as geometric objects that bear a deformation content. By refining the resolution of the surface deformation gradient afforded by the polar decomposition theorem, we identify a bending content and a class of deformations that leave it unchanged. These are the bending-neutral deformations, ful

Or Raveh, Junya Honda, Masashi Sugiyama

Various approaches have emerged for multi-armed bandits in distributed systems. The multiplayer dueling bandit problem, common in scenarios with only preference-based information like human feedback, introduces challenges related to controlling collaborative exploration of non-informative arm pairs, but has received little attention. To fill this gap, we dem

Marco Longinetti, Simone Naldi

A reformulation of the three circles theorem of Johnson with distance coordinates to the vertices of a triangle is explicitly represented in a polynomial system and solved by symbolic computation. A similar polynomial system in distance coordinates to the vertices of a tetrahedron $T \subset \mathbb{R}^3$ is introduced to represent the configurations of four