October 2023 arXiv papers — page 103

Jia-Wei Liu, Yan-Pei Cao, Jay Zhangjie Wu, Weijia Mao

Despite recent progress in diffusion-based video editing, existing methods are limited to short-length videos due to the contradiction between long-range consistency and frame-wise editing. Prior attempts to address this challenge by introducing video-2D representations encounter significant difficulties with large-scale motion- and view-change videos, espec

Leonardo F. R. Ribeiro, Mohit Bansal, Markus Dreyer

Readability refers to how easily a reader can understand a written text. Several factors affect the readability level, such as the complexity of the text, its subject matter, and the reader's background knowledge. Generating summaries based on different readability levels is critical for enabling knowledge consumption by diverse audiences. However, current t

W. A. van Wijngaarden, W. Happer

For 2n-stream radiation transfer theory, a stack of m clouds can be represented as an equivalent cloud. Individual clouds, indexed by c = 1, 2, 3, ..., m are characterized by 2n x 2n scattering matrices S^{c}, that describe how the cloud interacts with 2n streams of axially symmetric incoming radiation, propagating in upward and downward Gauss-Legendre sampl

M. K. Hooda, A. Chakraborty, S. Roy, R. Swami

We report a comprehensive study on the electronic transport properties of SrZn2Ge2 single crystals. The electrical resistivity of the compound exhibits metallic behavior, following a T^2 dependence below 35 K, consistent with the Fermi liquid behavior. However, a notable deviation is observed from this behavior at lower temperatures as a pronounced resistivi

V . D. Kuznetsov, A. I. Osin

The evolutionarity conditions for the MHD shock waves are considered within the framework of the 8-moment approximation for collisionless plasma with heat fluxes. In the general case, evolutionarity diagrams are obtained depending on the relative magnitude of the Alfven wave velocity in front of or behind the shock wave front. The evolutionarity conditions f

Marco Rauscher, Alessandro Scagliotti, Felipe Pagginelli Patricio

In this paper, we consider the signature-to-path reconstruction problem from the control theoretic perspective. Namely, we design an optimal control problem whose solution leads to the minimal-length path that generates a given signature. In order to do that, we minimize a cost functional consisting of two competing terms, i.e., a weighted final-time cost co

Petar Mlinarić, Peter Benner, Serkan Gugercin

Interpolatory necessary optimality conditions for $\mathcal{H}_2$-optimal reduced-order modeling of unstructured linear time-invariant (LTI) systems are well-known. Based on previous work on $\mathcal{L}_2$-optimal reduced-order modeling of stationary parametric problems, in this paper we develop and investigate optimality conditions for $\mathcal{H}_2$-opti

Taylor Daniels

For $n \in \mathbb{N}$ let $\Pi[n]$ denote the set of partitions of $n$, i.e., the set of positive integer tuples $(x_1,x_2,\ldots,x_k)$ such that $x_1 \geq x_2 \geq \cdots \geq x_k$ and $x_1 + x_2 + \cdots + x_k = n$. Fixing $f:\mathbb{N}\to\{0,\pm 1\}$, for $\pi = (x_1,x_2,\ldots,x_k) \in \Pi[n]$ let $f(\pi) := f(x_1)f(x_2)\cdots f(x_k)$. In this way we de

Tianyu Guo, Wei Hu, Song Mei, Huan Wang

While large language models based on the transformer architecture have demonstrated remarkable in-context learning (ICL) capabilities, understandings of such capabilities are still in an early stage, where existing theory and mechanistic understanding focus mostly on simple scenarios such as learning simple function classes. This paper takes initial steps on

Rob Behary, Irina Novikova, Eugeniy Mikhailov, Alex Gill

The sensitivity of electric field sensors based on two-photon electromagnetically induced transparency (EIT) involving highly excited Rydberg states in thermal atoms is often limited by the residual Doppler effect and optical power broadening. Here, we propose a method to reduce the EIT spectral linewidth using a Ramsey interrogation approach, allowing multi

Jiajie Kong, Tony Pourmohamad, Herbert K. H. Lee

Bayesian optimization (BO) developed as an approach for the efficient optimization of expensive black-box functions without gradient information. A typical BO paper introduces a new approach and compares it to some alternatives on simulated and possibly real examples to show its efficacy. Yet on a different example, this new algorithm might not be as effecti

Qian Feng, Sing Kiong Nguang

This paper examines the problem of stabilizing linear distributed delay systems with nonlinear distributed delay kernels and dissipativity constraints. Specifically, the nonlinear distributed kernel includes functions such as polynomials, trigonometric and exponential functions. By constructing a Liapunov-Krasovski\v{i} functional related to the distributed

Arturo Del Castillo Bernal, Philippe Decoste, James Richard Forbes

This paper considers homography estimation in a Bayesian filtering framework using rate gyro and camera measurements. The use of rate gyro measurements facilitates a more reliable estimate of homography in the presence of occlusions, while a Bayesian filtering approach generates both a homography estimate along with an uncertainty. Uncertainty information op

Taejong Joo, Diego Klabjan

Distribution shifts pose significant challenges for model calibration and model selection tasks in the unsupervised domain adaptation problem -- a scenario where the goal is to perform well in a distribution shifted domain without labels. In this work, we tackle difficulties coming from distribution shifts by developing a novel importance weighted group accu

Jerry Zhi-Yang He, Zackory Erickson, Daniel S. Brown, Anca D. Dragan

Our ultimate goal is to build robust policies for robots that assist people. What makes this hard is that people can behave unexpectedly at test time, potentially interacting with the robot outside its training distribution and leading to failures. Even just measuring robustness is a challenge. Adversarial perturbations are the default, but they can paint th

Taylor Daniels

For $n \in \mathbb{N}$ let $\Pi[n]$ denote the set of partitions of $n$, i.e., the set of positive integer tuples $(x_1,x_2,\ldots,x_k)$ such that $x_1 \geq x_2 \geq \cdots \geq x_k$ and $x_1 + x_2 + \cdots + x_k = n$. Fixing $f:\mathbb{N}\to\{0,\pm 1\}$, for $\pi = (x_1,x_2,\ldots,x_k) \in \Pi[n]$ let $f(\pi) := f(x_1)f(x_2)\cdots f(x_k)$. In this way we de

Rallou A. Chatzimichail, Aristides T. Hatjimihail

Although there is extensive literature on the application of artificial neural networks (NNs) in quality control (QC), to monitor the conformity of a process to quality specifications, at least five QC measurements are required, increasing the related cost. To explore the application of neural networks to samples of QC measurements of very small size, four o

Tianle Huang, Nitish Sontakke, K. Niranjan Kumar, Irfan Essa

Domain randomization (DR), which entails training a policy with randomized dynamics, has proven to be a simple yet effective algorithm for reducing the gap between simulation and the real world. However, DR often requires careful tuning of randomization parameters. Methods like Bayesian Domain Randomization (Bayesian DR) and Active Domain Randomization (Adap

Bo Ni, David L. Kaplan, Markus J. Buehler

Through evolution, nature has presented a set of remarkable protein materials, including elastins, silks, keratins and collagens with superior mechanical performances that play crucial roles in mechanobiology. However, going beyond natural designs to discover proteins that meet specified mechanical properties remains challenging. Here we report a generative

Dimitrios Bralios, Gordon Wichern, François G. Germain, Zexu Pan

The introduction of audio latent diffusion models possessing the ability to generate realistic sound clips on demand from a text description has the potential to revolutionize how we work with audio. In this work, we make an initial attempt at understanding the inner workings of audio latent diffusion models by investigating how their audio outputs compare w

Chendi Qian, Didier Chételat, Christopher Morris

Recently, machine learning, particularly message-passing graph neural networks (MPNNs), has gained traction in enhancing exact optimization algorithms. For example, MPNNs speed up solving mixed-integer optimization problems by imitating computational intensive heuristics like strong branching, which entails solving multiple linear optimization problems (LPs)

Tariq Alkhalifah, Xinquan Huang

Recently, Physics-Informed Neural Networks (PINNs) have gained significant attention for their versatile interpolation capabilities in solving partial differential equations (PDEs). Despite their potential, the training can be computationally demanding, especially for intricate functions like wavefields. This is primarily due to the neural-based (learned) ba

Yin-Kai Yu, Zhi Zeng, Yu-Rong Shu, Zi-Xiang Li

Quantum criticality within Dirac fermions harbors a plethora of exotic phenomena, attracting sustained attention in the past decades. Here, we explore the imaginary-time relaxation dynamics in a typical Dirac quantum criticality belonging to chiral Heisenberg universality class. Performing large-scale quantum Monte Carlo simulation, we unveil rich nonequilib

Yuan Liu, Ho Yiu Chung, Emmanuel Zambrini Cruzeiro, Junior R. Gonzales-Ureta

We show that a quantum correlation $p$ is in a face of the nonsignaling polytope with no local points if and only if $p$ has nonlocal content 1, if and only if $p$ allows for a Greenberger-Horne-Zeilinger-like proof, and if and only if $p$ provides a perfect strategy for a nonlocal game. That is, face nonsignaling (FNS) correlations, full nonlocality (FN), a

Oscar Finegan

In this paper we produce the first known formula for cohomologies of the derived tensor products of structure sheaves of subschemes in the case where the intersection of the subschemes is not a local complete intersection. The case covered here is where the intersection instead consists of two local complete intersection components, one of codimension 1 and

Jenna C. Fromer, David E. Graff, Connor W. Coley

The discovery of therapeutic molecules is fundamentally a multi-objective optimization problem. One formulation of the problem is to identify molecules that simultaneously exhibit strong binding affinity for a target protein, minimal off-target interactions, and suitable pharmacokinetic properties. Inspired by prior work that uses active learning to accelera

Martin Scheiber, Alessandro Fornasier, Christian Brommer, Stephan Weiss

In this work, we explore the recent advances in equivariant filtering for inertial navigation systems to improve state estimation for uncrewed aerial vehicles (UAVs). Traditional state-of-the-art estimation methods, e.g., the multiplicative Kalman filter (MEKF), have some limitations concerning their consistency, errors in the initial state estimate, and con

Joao Vitor Parada Poletto, David Andrade, Marcelo V. Flamarion, Roberto Ribeiro-Jr

We present a novel numerical method for simulating the generation and propagation of surface gravity waves by vertical seabed displacements. The cornerstone of our method is the computation of a time dependent conformal map which incorporates the time dependent geometry of the seabed and the wave profile along the free surface. This enables us to handle gene

Stephen Cantrell, Eduardo Reyes

We compare the marked length spectra of some pairs of proper and cocompact cubical actions of a non-virtually cyclic group on $\text{CAT}(0)$ cube complexes. The cubulations are required to be virtually co-special, have the same sets of convex-cocompact subgroups, and admit a contracting element. There are many groups for which these conditions are always fu

Karim Radouane, Andon Tchechmedjiev, Julien Lagarde, Sylvie Ranwez

In this paper, we investigate building a sequence to sequence architecture for motion to language translation and synchronization. The aim is to translate motion capture inputs into English natural-language descriptions, such that the descriptions are generated synchronously with the actions performed, enabling semantic segmentation as a byproduct, but witho

O. Fedchenko, F. K. Diekmann, P. Russmann, M. Kallmayer

A comprehensive study of the electronic and structural phase transition from 1T` to Td in the bulk Weyl semimetal Mo1-xWxTe2 at different doping concentrations has been carried out using time-of-flight momentum microscopy (including circular and linear dichroism), X-ray photoelectron spectroscopy (XPS), X-ray photoelectron diffraction (XPD), X-ray diffractio

Osama A. Marzouk

The collision cross sections (CCS), momentum transfer cross sections (MTCS), or scattering cross sections (SCS) of an electron neutral pair are important components for computing the electric conductivity of a plasma gas. Larger collision cross sections for electrons moving freely within neutral particles (molecules or atoms) cause more scattering of these e

Ivy I. Chen, Jennifer Solgaard, Ryoto Sekine, Azmain A. Hossain

Lithium niobate (LiNbO$_3$, LN) is a ferroelectric crystal of interest for integrated photonics owing to its large second-order optical nonlinearity and the ability to impart periodic poling via an external electric field. However, on-chip device performance based on thin-film lithium niobate (TFLN) is presently limited by propagation losses arising from sur

Haozhe Chen, Junfeng Yang, Carl Vondrick, Chengzhi Mao

Large-scale pre-trained vision foundation models, such as CLIP, have become de facto backbones for various vision tasks. However, due to their black-box nature, understanding the underlying rules behind these models' predictions and controlling model behaviors have remained open challenges. We present a framework for interpreting vision transformer's latent

Ji Qi, Kaixuan Ji, Xiaozhi Wang, Jifan Yu

Open Information Extraction (OIE) aims to extract objective structured knowledge from natural texts, which has attracted growing attention to build dedicated models with human experience. As the large language models (LLMs) have exhibited remarkable in-context learning capabilities, a question arises as to whether the task of OIE can be effectively tackled w

Michelle C. Jecmen, M. S. Oey

The classical model of massive-star mechanical feedback is based on effects at solar metallicity (Zsun), yet feedback parameters are very different at low metallicity. Metal-poor stellar winds are much weaker, and more massive supernova progenitors likely collapse directly to black holes without exploding. Thus, for ~0.4 Zsun we find reductions in the total

Pavel Krupskii, Raphaël Huser

In this paper, we introduce a new class of models for spatial data obtained from max-convolution processes based on indicator kernels with random shape. We show that this class of models have appealing dependence properties including tail dependence at short distances and independence at long distances. We further consider max-convolutions between such proce

Matthew R. Oster, Ilya Amburg, Samrat Chatterjee, Daniel A. Eisenberg

Resilient operation of interdependent infrastructures against compound hazard events is essential for maintaining societal well-being. To address consequence assessment challenges in this problem space, we propose a novel tri-level optimization model applied to a proof-of-concept case study with fuel distribution and transportation networks -- encompassing o

Ji Qi, Kaixuan Ji, Jifan Yu, Duokang Wang

Building models that comprehends videos and responds specific user instructions is a practical and challenging topic, as it requires mastery of both vision understanding and knowledge reasoning. Compared to language and image modalities, training efficiency remains a serious problem as existing studies train models on massive sparse videos paired with brief

Joris Verhagen, Lars Lindemann, Jana Tumova

Signal Temporal Logic (STL) is a formal language over continuous-time signals (such as trajectories of a multi-agent system) that allows for the specification of complex spatial and temporal system requirements (such as staying sufficiently close to each other within certain time intervals). To promote robustness in multi-agent motion planning with such comp

Chenxu Zhao, Wei Qian, Yucheng Shi, Mengdi Huai

Deep neural networks have exhibited remarkable performance across a wide range of real-world tasks. However, comprehending the underlying reasons for their effectiveness remains a challenging problem. Interpreting deep neural networks through examining neurons offers distinct advantages when it comes to exploring the inner workings of neural networks. Previo

Ugur Cetiner, Jeremy Gunawardena

We consider a finite-state, continuous-time Markov process, represented in the "linear framework" by a directed graph with labelled edges which specifies the infinitesimal generator of the process. If the graph is strongly connected, the process has a unique steady-state probability distribution, $p$, which may not be one of thermodynamic equilibrium. If the

Sagi Shaier, Lawrence E. Hunter, Katharina von der Wense

Both standalone language models (LMs) as well as LMs within downstream-task systems have been shown to generate statements which are factually untrue. This problem is especially severe for low-resource languages, where training data is scarce and of worse quality than for high-resource languages. In this opinion piece, we argue that LMs in their current stat

Tristan Benoist, Laurent Bruneau, Vojkan Jaksic, Annalisa Panati

Recent theoretical investigations of the two-times measurement entropy production (2TMEP) in quantum statistical mechanics have shed a new light on the mathematics and physics of the quantum-mechanical probabilistic rules. Among notable developments are the extensions of entropic fluctuation relations to quantum domain and discovery of a deep link between 2T

David Burns, Daniel Macias Castillo, Soogil Seo

The theory of Weil-Stark elements is used to develop an axiomatic approach to the formulation of refined versions of Stark's Conjecture. This gives concrete new results concerning leading terms of Artin $L$-series and arithmetic properties of Stark elements.

Dolores Martín Barquero, Cándido Martín González, Iván Ruiz Campos

The theory of path algebras is usually circunscripted to the study of representations, usually linked to finite graphs. In our work, we focus on studying the structure of path algebras over a field associated to arbitrary graphs. We characterise perfection (simplicity, primitivity, primeness and semiprimeness) and finitness conditions (artinianity, semiartin

Huey-Wen Lin

We present the first lattice-QCD $x$-dependent pion valence-quark generalized parton distribution (GPD) calculated directly at physical pion mass using the Large-Momentum Effective Theory (LaMET) with next-to-next-to-leading order perturbative matching correction. We use clover fermions for the valence action on $2+1+1$ flavors of highly improved staggered q

Ruichen Wang, Samuel Audia, Dinesh Manocha

We present a novel algorithm that enhances the accuracy of electromagnetic field simulations in indoor environments by incorporating the Uniform Geometrical Theory of Diffraction (UTD) for surface diffraction. This additional diffraction phenomenology is important for the design of modern wireless systems and allows us to capture the effects of more complex

Mouhamed Moustapha Fall, Tobias Weth

We prove that positive solutions $u\in H^s(\mathbb{R}^N)$ to the equation $(-\Delta )^s u+ u=u^p$ in $\mathbb{R}^N$ are nonradially nondegenerate, for all $s\in (0,1)$, $N\geq 1$ and $p>1$ strictly smaller than the critical Sobolev exponent. By this we mean that the linearized equation $(-\Delta )^s w+ w-pu^{p-1}w = 0$ does not admit nonradial solutions besi

Samuele Maschio, Alexandre Miquel

In this paper we show that using implicative algebras one can produce models of set theory generalizing Heyting/Boolean-valued models and realizability models of (I)ZF, both in intuitionistic and classical logic. This has as consequence that any topos which is obtained from a Set-based tripos as the result of the tripos-to-topos construction hosts a model of

Ruxandra Barbulescu, Tiago Marques, Arlindo L. Oliveira

While some convolutional neural networks (CNNs) have achieved great success in object recognition, they struggle to identify objects in images corrupted with different types of common noise patterns. Recently, it was shown that simulating computations in early visual areas at the front of CNNs leads to improvements in robustness to image corruptions. Here, w

Mohsen Hamedi, Brian Vermeire

We investigate the feasibility of gradient-free aeroacoustic shape optimization using the Flux Reconstruction (FR) approach to study two-dimensional flow at low Reynolds numbers. The Overall Sound Pressure Level (OASPL) is computed via the direct acoustic approach, and optimization is performed using the gradient-free Mesh Adaptive Direct Search (MADS) algor

Cynthia Dwork, Chris Hays, Jon Kleinberg, Manish Raghavan

We study the impact of content moderation policies in online communities. In our theoretical model, a platform chooses a content moderation policy and individuals choose whether or not to participate in the community according to the fraction of user content that aligns with their preferences. The effects of content moderation, at first blush, might seem obv

Richard Lechner, Thomas Speckhofer

A Haar system Hardy space is the completion of the linear span of the Haar system $(h_I)_I$, either under a rearrangement-invariant norm $\|\cdot \|$ or under the associated square function norm \begin{equation*} \Bigl\| \sum_Ia_Ih_I \Bigr\|_{*} = \Bigl\| \Bigl( \sum_I a_I^2 h_I^2 \Bigr)^{1/2} \Bigr\|. \end{equation*} Apart from $L^p$, $1\le p<\infty$, the c

Sagi Shaier, Kevin Bennett, Lawrence Hunter, Katharina von der Wense

State-of-the-art question answering (QA) models exhibit a variety of social biases (e.g., with respect to sex or race), generally explained by similar issues in their training data. However, what has been overlooked so far is that in the critical domain of biomedicine, any unjustified change in model output due to patient demographics is problematic: it resu

Mathieu Ravaut, Aixin Sun, Nancy F. Chen, Shafiq Joty

Large language models (LLMs) excel in abstractive summarization tasks, delivering fluent and pertinent summaries. Recent advancements have extended their capabilities to handle long-input contexts, exceeding 100k tokens. However, in question answering, language models exhibit uneven utilization of their input context. They tend to favor the initial and final

A. H. Ajjath, Hua-Sheng Shao

In this talk, we discuss the soft-gluon resummation for a pair of Higgs bosons in the dominant gluon fusion channel to next-to-next-to-next-to-leading logarithmic (N$^3$LL) accuracy. Through the study, we achieve sub-percent level accuracy in the uncertainties from the residual renormalisation and factorisation scales, for both inclusive and differential mas

Alina Bucur, Francesc Fité, Kiran S. Kedlaya

Let A and A' be nonzero abelian varieties defined over a number field k such that Hom(A,A')=0. Under the Generalized Riemann hypothesis for motivic L-functions attached to A and A', we show that there exists a prime p of k of good reduction for A and A' at which the Frobenius traces of A and A' are nonzero and differ by sign, and such that the norm of p is O

Jingcheng Deng, Liang Pang, Huawei Shen, Xueqi Cheng

Retrieval-augmented language models show promise in addressing issues like outdated information and hallucinations in language models (LMs). However, current research faces two main problems: 1) determining what information to retrieve, and 2) effectively combining retrieved information during generation. We argue that valuable retrieved information should n

Boštjan Brešar, Arti Pandey, Gopika Sharma

A sequence $S$ of vertices of a graph $G$ is called a dominating sequence of $G$ if $(i)$ each vertex $v$ of $S$ dominates a vertex of $G$ that was not dominated by any of the vertices preceding vertex $v$ in $S$, and $(ii)$ every vertex of $G$ is dominated by at least one vertex of $S$. The Grundy Domination problem is to find a longest dominating sequence

Lanxiang Xing, Haixu Wu, Yuezhou Ma, Jianmin Wang

Fluid prediction is a long-standing challenge due to the intrinsic high-dimensional non-linear dynamics. Previous methods usually utilize the non-linear modeling capability of deep models to directly estimate velocity fields for future prediction. However, skipping over inherent physical properties but directly learning superficial velocity fields will overw

Alexios A. Michailidis, Dmitry A. Abanin, Luca V. Delacrétaz

Transport and the approach to equilibrium in interacting classical and quantum systems is a challenging problem of both theoretical and experimental interest. One useful organizing principle characterizing equilibration is the dissipative universality class, the most prevalent one being diffusion. In this paper, we use the effective field theory (EFT) of dif

Zhicheng Cai, Xiaohan Ding, Qiu Shen, Xun Cao

We propose Re-parameterized Refocusing Convolution (RefConv) as a replacement for regular convolutional layers, which is a plug-and-play module to improve the performance without any inference costs. Specifically, given a pre-trained model, RefConv applies a trainable Refocusing Transformation to the basis kernels inherited from the pre-trained model to esta

Sang-Wook Cheong, Fei-Ting Huang

Ferromagnetism can be characterized by various unique phenomena such as non-zero magnetization (inducing magnetic attraction/repulsion), diagonal piezomagnetism, nonreciprocal circular dichroism (such as Faraday effect), odd-order (including linear) anomalous Hall effect, and magneto-optical Kerr effect. We identify all broken symmetries requiring each of th

Zhuohao Liu, Emma C. Johnson, David L. Feder

All known resource states for measurement-based quantum teleportation in correlation space possess symmetry protected topological order, but is this a sufficient or even necessary condition? This work considers two families of one-dimensional qubit states to answer this question in the negative. The first is a family of matrix-product states with bond dimens

Animesh Basak Chowdhury, Shailja Thakur, Hammond Pearce, Ramesh Karri

Despite the growing interest in ML-guided EDA tools from RTL to GDSII, there are no standard datasets or prototypical learning tasks defined for the EDA problem domain. Experience from the computer vision community suggests that such datasets are crucial to spur further progress in ML for EDA. Here we describe our experience curating two large-scale, high-qu

Mouad El Bouchattaoui, Myriam Tami, Benoit Lepetit, Paul-Henry Cournède

Accurately estimating treatment effects over time is crucial in fields such as precision medicine, epidemiology, economics, and marketing. Many current methods for estimating treatment effects over time assume that all confounders are observed or attempt to infer unobserved ones. In contrast, our approach focuses on unobserved adjustment variables, which spe

Yue Xia, Lijuan Chen, Vaibhava Srivastava, Rana D. Parshad

In the current manuscript, a first two-patch model with Allee effect and nonlinear dispersal is presented. We study both the ODE case and the PDE case here. In the ODE model, the stability of the equilibrium points and the existence of saddle-node bifurcation are discussed. The phase diagram and bifurcation curve of our model are also given by numerical simu

Sierra Seacat, John L. Lyons, Hartwin Peelaers

Ga$_2$O$_3$ is a wide-bandgap material of interest for a wide variety of devices, many of these requiring heterostructures, for instance to achieve carrier confinement. A common method to create such heterostructures is to alloy with In$_2$O$_3$ or Al$_2$O$_3$. However, the lattice constants of these materials are significantly different from those of Ga$_2$

Zihao Li, Xiang Ji, Minshuo Chen, Mengdi Wang

A recently popular approach to solving reinforcement learning is with data from human preferences. In fact, human preference data are now used with classic reinforcement learning algorithms such as actor-critic methods, which involve evaluating an intermediate policy over a reward learned from human preference data with distribution shift, known as off-polic

W. Lin, K. Worden, E. J. Cross

An important challenge faced by wind farm operators is to reduce operation and maintenance cost. Structural health monitoring provides a means of cost reduction through minimising unnecessary maintenance trips as well as prolonging turbine service life. Population-based structural health monitoring can further reduce the cost of health monitoring systems by

Bakhtiyor Narzilloev, Ahmadjon Abdujabbarov, Bobomurat Ahmedov, Cosimo Bambi

In this work, we investigate the electromagnetic energy released by astrophysical black holes within the Kerr-Taub-NUT solution, which describes rotating black holes with a nonvanishing gravitomagnetic charge. In our study, we consider the black holes in the X-ray binary systems GRS 1915+105, GRO J1655-40, XTE J1550-564, A0620-00, H1743-322, and GRS 1124-683

Zhe Wang, Petar Veličković, Daniel Hennes, Nenad Tomašev

Identifying key patterns of tactics implemented by rival teams, and developing effective responses, lies at the heart of modern football. However, doing so algorithmically remains an open research challenge. To address this unmet need, we propose TacticAI, an AI football tactics assistant developed and evaluated in close collaboration with domain experts fro

Javier de Frutos, Bosco Garcia-Archilla, Julia Novo

In this paper we consider the numerical approximation of infinite horizon problems via the dynamic programming approach. The value function of the problem solves a Hamilton-Jacobi-Bellman (HJB) equation that is approximated by a fully discrete method. It is known that the numerical problem is difficult to handle by the so called curse of dimensionality. To m

Anirudh Som, Karan Sikka, Helen Gent, Ajay Divakaran

Paraphrasing of offensive content is a better alternative to content removal and helps improve civility in a communication environment. Supervised paraphrasers; however, rely heavily on large quantities of labelled data to help preserve meaning and intent. They also often retain a large portion of the offensiveness of the original content, which raises quest

Yuheng Yang, Duanduan Wan

We have discovered that two significant quantities within hard particle systems: the probability of successfully inserting an additional particle at random and the scale distribution function, can be connected by a concise relation. We anticipate that this relation holds universal applicability for convex hard particles. Our investigations encompassed a rang

Dung Truong, Muhammad Abdullah Khalid, Arnaud Delorme

The ability of Deep Learning to process and extract relevant information in complex brain dynamics from raw EEG data has been demonstrated in various recent works. Deep learning models, however, have also been shown to perform best on large corpora of data. When processing EEG, a natural approach is to combine EEG datasets from different experiments to train

Mai Le, Thien Huynh-The, Tan Do-Duy, Thai-Hoc Vu

The emergence of new services and applications in emerging wireless networks (e.g., beyond 5G and 6G) has shown a growing demand for the usage of artificial intelligence (AI) in the Internet of Things (IoT). However, the proliferation of massive IoT connections and the availability of computing resources distributed across future IoT systems have strongly de

Roman Dautzenberg, Timo Küster, Timon Mathis, Yann Roth

Drilling, grinding, and setting anchors on vertical walls are fundamental processes in everyday construction work. Manually doing these works is error-prone, potentially dangerous, and elaborate at height. Today, heavy mobile ground robots can perform automatic power tool work. However, aerial vehicles could be deployed in untraversable environments and reac

Seunggeun Chi, Hyung-gun Chi, Qixing Huang, Karthik Ramani

Skeleton-based action recognition has made significant advancements recently, with models like InfoGCN showcasing remarkable accuracy. However, these models exhibit a key limitation: they necessitate complete action observation prior to classification, which constrains their applicability in real-time situations such as surveillance and robotic systems. To o

David Criens, Lars Niemann

In this paper we study a family of nonlinear (conditional) expectations that can be understood as a semimartingale with uncertain local characteristics. Here, the differential characteristics are prescribed by a time and path-dependent set-valued function. We show that the associated control problem coincides with both its weak and relaxed counterparts. Furt

Xin Bing, Xin He, Dian Jin, Yuqian Zhang

Vintage factor analysis is one important type of factor analysis that aims to first find a low-dimensional representation of the original data, and then to seek a rotation such that the rotated low-dimensional representation is scientifically meaningful. The most widely used vintage factor analysis is the Principal Component Analysis (PCA) followed by the va

Laurence T Maloney, Maria F Dal Martello, Vivian Fei, Valerie Ma

English speakers use probabilistic phrases such as likely to communicate information about the probability or likelihood of events. Communication is successful to the extent that the listener grasps what the speaker means to convey and, if communication is successful, individuals can potentially coordinate their actions based on shared knowledge about uncert

Hassan Shahmohammadi, Adhiraj Ghosh, Hendrik P. A. Lensch

Figurative and non-literal expressions are profoundly integrated in human communication. Visualising such expressions allow us to convey our creative thoughts, and evoke nuanced emotions. Recent text-to-image models like Stable Diffusion, on the other hand, struggle to depict non-literal expressions. Recent works primarily deal with this issue by compiling h

Alisher Duspayev, Ryan Cardman, David A. Anderson, Georg Raithel

We prepare and analyze Rydberg states with orbital quantum numbers $\ell \le 6$ using three-optical-photon electromagnetically-induced transparency (EIT) and radio-frequency (RF) dressing, and employ the high-$\ell$ states in electric-field sensing. Rubidium-85 atoms in a room-temperature vapor cell are first promoted into the $25F_{5/2}$ state via Rydberg-E

Jiyuan Shen, Wenzhuo Yang, Kwok-Yan Lam

Training large AI models typically requires large-scale datasets in the machine learning process, making training and parameter-tuning process both time-consuming and costly. Some researchers address this problem by carefully synthesizing a very small number of highly representative and informative samples from real-world datasets. This approach, known as Da

Chengyun Hua, David A. Tennant, Andrei Savici, Vladislav Sedov

Exciting new fundamental scientific questions are currently being raised regarding nonequilibrium dynamics in spin systems, as this directly relates to low power and low loss energy transport for spintronics. Inelastic neutron scattering (INS) is an indispensable tool to study spin excitations in complex magnetic materials. However, conventional INS spectrom

Harrison Chen, Gurbir Dhillon

Let $G$ and $\check{G}$ be Langlands dual connected reductive groups. We establish a monoidal equivalence of $\infty$-categories between equivariant quasicoherent sheaves on the formal neighborhood of the nilpotent cone in $G$ and Steinberg-Whittaker D-modules on the loop group of $\check{G}$, as conjectured by Bezrukavnikov. More generally, we establish equ

Ananda Roy, Sameer Erramilli, Robert M. Konik

Efficient preparation of arbitrary entangled quantum states is crucial for quantum computation. This is particularly important for noisy intermediate scale quantum simulators relying on variational hybrid quantum-classical algorithms. To that end, we propose symmetry-conserving modified quantum approximate optimization algorithm~(SCom-QAOA) circuits. The dep

Bita Darvish Rouhani, Ritchie Zhao, Ankit More, Mathew Hall

Narrow bit-width data formats are key to reducing the computational and storage costs of modern deep learning applications. This paper evaluates Microscaling (MX) data formats that combine a per-block scaling factor with narrow floating-point and integer types for individual elements. MX formats balance the competing needs of hardware efficiency, model accur

V. H. Mikaelian

Explicit embeddings of the group $\mathbb{Q}$ into a finitely presented group $\mathcal{Q}$ and into a $2$-generator finitely presented group $T_{\mathcal{Q}}$ are suggested. The constructed embeddings reflect questions mentioned by Johnson, Bridson, de la Harpe in the literature in late $1990$s about possibility of such embeddings for $\mathbb{Q}$. Techniqu

Davor Dragičević, Xiao Tang, Wenmeng Zhang

Takens Theorem for a partially hyperbolic dynamics provides a normal linearization along the center manifold. In this paper, we give the nonautonomous version of Takens Theorem under non-resonance conditions formulated in terms of the dichotomy spectrum. In our proof, one difficulty is to solve homological equations for the normal form theory which involve a

Fredrik Hellström, Benjamin Guedj

We derive generic information-theoretic and PAC-Bayesian generalization bounds involving an arbitrary convex comparator function, which measures the discrepancy between the training and population loss. The bounds hold under the assumption that the cumulant-generating function (CGF) of the comparator is upper-bounded by the corresponding CGF within a family

Wentong Li, Yuqian Yuan, Song Wang, Wenyu Liu

Weakly-supervised segmentation with label-efficient sparse annotations has attracted increasing research attention to reduce the cost of laborious pixel-wise labeling process, while the pairwise affinity modeling techniques play an essential role in this task. Most of the existing approaches focus on using the local appearance kernel to model the neighboring

Fabian David Schmidt, Ivan Vulić, Goran Glavaš

Multilingual language models enable zero-shot cross-lingual transfer (ZS-XLT): fine-tuned on sizable source-language task data, they perform the task in target languages without labeled instances. The effectiveness of ZS-XLT hinges on the linguistic proximity between languages and the amount of pretraining data for a language. Because of this, model selectio

Albert Alonso, Julius B. Kirkegaard

We investigate the boundary between chemotaxis driven by spatial estimation of gradients and chemotaxis driven by temporal estimation. While it is well known that spatial chemotaxis becomes disadvantageous for small organisms at high noise levels, it is unclear whether there is a discontinuous switch of optimal strategies or a continuous transition exists. H

Antoine Van Biesbroeck

In Bayesian theory, the role of information is central. The influence exerted by prior information on posterior outcomes often jeopardizes Bayesian studies, due to the potentially subjective nature of the prior choice. In modeling where a priori knowledge is lacking, the reference prior theory emerges as a proficient tool. Based on the criterion of mutual in

Carlo Mercuri, Riccardo Molle

In this paper we consider nonlinear elliptic PDEs of the type $$-\Delta_p u+a(x)|u|^{p-2}u=|u|^{p^*-2}u \qquad \mbox{ in }\Omega,$$ where $1<p<N$ and $p^*=Np/(N-p)$ is the critical Sobolev exponent, and allowing the asymptotic behavior of the weight function $a$ to be sensitive to the direction. We provide a unified variational approach to obtain existence o

Jiri Podolsky, Matus Papajcik

A new effective approach to the algebraic classification of geometries in 2+1 gravity is presented. It uses five real Cotton scalars $\Psi_A$ of distinct boost weights, which are 3D analogues of the Newman-Penrose scalars representing the Weyl tensor in 4D. The classification into types I, II, D, III, N, O is directly related to the multiplicity of the four

Hongliang Chen, Guangchen Su, Xin Fu, Lin Yang

The 2$\times$2 3-dB couplers are one of the most widely used and important components in silicon photonics. We propose an ultra-broadband and compact 2$\times$2 3-dB adiabatic coupler defined by b-splines and optimized with an efficient supermode-injected adjoint shape optimization. By employing mode adiabatic evolution and mode coupling at two different wav