September 2022 arXiv papers

Sayma Sultana, Asif Kamal Turzo, Amiangshu Bosu

Context: Contemporary software development organizations lack diversity and the ratios of women in Free and open-source software (FOSS) communities are even lower than the industry average. Although the results of recent studies hint the existence of biases against women, it is unclear to what extent such biases influence the outcomes of various software dev

Nikola Paunkovic, Marko Vojinovic

We give a general overview of various flavors of the equivalence principle in classical and quantum physics, with special emphasis on the so-called weak equivalence principle, and contrast its validity in mechanics versus field theory. We also discuss its generalisation to a theory of quantum gravity. Our analysis suggests that only the strong equivalence pr

Yizhou Zhao, Zhenyang Li, Xun Guo, Yan Lu

Temporal modeling is crucial for various video learning tasks. Most recent approaches employ either factorized (2D+1D) or joint (3D) spatial-temporal operations to extract temporal contexts from the input frames. While the former is more efficient in computation, the latter often obtains better performance. In this paper, we attribute this to a dilemma betwe

Emily McMilin

Modern language modeling tasks are often underspecified: for a given token prediction, many words may satisfy the user's intent of producing natural language at inference time, however only one word will minimize the task's loss function at training time. We introduce a simple causal mechanism to describe the role underspecification plays in the gene

Mario Novello, Vicente Antunes

Despite the success of the Higgs mechanism to account for the generation of the masses of Standard Model (SM) elementary particles, the ultimate nature and origin of "mass" remain open questions in contemporary physics. From a foundational perspective, mass should be fundamentally related to the gravitational interaction and, according to Mach, to th

Honglin Chen, Rundi Wu, Eitan Grinspun, Changxi Zheng

Implicit Neural Spatial Representation (INSR) has emerged as an effective representation of spatially-dependent vector fields. This work explores solving time-dependent PDEs with INSR. Classical PDE solvers introduce both temporal and spatial discretizations. Common spatial discretizations include meshes and meshless point clouds, where each degree-of-freedo

Pankaj K. Agarwal, Tzvika Geft, Dan Halperin, Erin Taylor

We study the problem of motion planning for a collection of $n$ labeled unit disc robots in a polygonal environment. We assume that the robots have revolving areas around their start and final positions: that each start and each final is contained in a radius $2$ disc lying in the free space, not necessarily concentric with the start or final position, which

Ruiqi Ni, Ahmed H. Qureshi

Neural Motion Planners (NMPs) have emerged as a promising tool for solving robot navigation tasks in complex environments. However, these methods often require expert data for learning, which limits their application to scenarios where data generation is time-consuming. Recent developments have also led to physics-informed deep neural models capable of repre

Yulun Wu, Robert A. Barton, Zichen Wang, Vassilis N. Ioannidis

Predicting the responses of a cell under perturbations may bring important benefits to drug discovery and personalized therapeutics. In this work, we propose a novel graph variational Bayesian causal inference framework to predict a cell's gene expressions under counterfactual perturbations (perturbations that this cell did not factually receive), levera

Sebastián Donoso, Alejandro Maass, Samuel Petite

We develop a geometric framework to address asymptoticity and nonexpansivity in topological dynamics. Our framework can be applied when the acting group is second countable and locally compact. As an application, we show extensions of Schwartzman's theorem in this context. Also, we get new results when the acting groups is ${\mathbb Z}^d$: any half-space

E. A. Huerta, Ben Blaiszik, L. Catherine Brinson, Kristofer E. Bouchard

A foundational set of findable, accessible, interoperable, and reusable (FAIR) principles were proposed in 2016 as prerequisites for proper data management and stewardship, with the goal of enabling the reusability of scholarly data. The principles were also meant to apply to other digital assets, at a high level, and over time, the FAIR guiding principles h

Peter B. Denton, Alessio Giarnetti, Davide Meloni

Next generation neutrino oscillation experiments are expected to measure the remaining oscillation parameters with very good precision. They will have unprecedented capabilities to search for new physics that modify oscillations. DUNE, with its broad band beam, good particle identification, and relatively high energies will provide an excellent environment t

Eleanor Clifford, Ilia Shumailov, Yiren Zhao, Ross Anderson

Early backdoor attacks against machine learning set off an arms race in attack and defence development. Defences have since appeared demonstrating some ability to detect backdoors in models or even remove them. These defences work by inspecting the training data, the model, or the integrity of the training procedure. In this work, we show that backdoors can

Chris Lin, Hugh Chen, Chanwoo Kim, Su-In Lee

Despite the widespread use of unsupervised models, very few methods are designed to explain them. Most explanation methods explain a scalar model output. However, unsupervised models output representation vectors, the elements of which are not good candidates to explain because they lack semantic meaning. To bridge this gap, recent works defined a scalar exp

Mohammed A. Chamma, Fereshteh Rajabi, Aishwarya Kumar, Martin Houde

We survey the spectro-temporal properties of fast radio bursts from FRB 20121102A observed by earlier studies across a wide range of frequencies. We investigate 167 bursts from FRB 20121102A spanning frequencies 1--7.5GHz, durations of less than 1 ms to approximately 10 ms, with low and high energies, and with different wait-times. We find from this sample o

Xiaotian Han, Tong Zhao, Yozen Liu, Xia Hu

Training graph neural networks (GNNs) on large graphs is complex and extremely time consuming. This is attributed to overheads caused by sparse matrix multiplication, which are sidestepped when training multi-layer perceptrons (MLPs) with only node features. MLPs, by ignoring graph context, are simple and faster for graph data, however they usually sacrifice

Diulhio Candido de Oliveira, Bogdan Tomoyuki Nassu, Marco Aurelio Wehrmeister

In this paper, we introduce an approach for detecting modifications in assembled printed circuit boards based on photographs taken without tight control over perspective and illumination conditions. One instance of this problem is the visual inspection of gas pumps PCBs, which can be modified by fraudsters wishing to deceive costumers or evade taxes. Given t

Xiaowen Feng, Heng Li

Background: In the metagenome assembly of a microbiome community, we may think abundant species would be easier to assemble due to their deeper coverage. However, this conjucture is rarely tested. We often do not know how many abundant species we are missing and do not have an approach to recover these species. Results: Here we proposed k-mer based and 16S R

Danrui Ni, Ranuri S. Dissanayaka Mudiyanselage, Xianghan Xu, Junsik Mun

A previously unreported layered spin 1/2 triangular lattice polymorph of TiI3 is described, synthesized under 6 GPa of applied pressure at 900 C, but stable at atmospheric pressure. This air-sensitive material has a CdI2-type layered structure (P-3m1 (#164), a = 4.012 A and c = 6.641 A at 120 K, Z = 1 of Ti0.667I2) with an in-plane triangular lattice, relate

Amin Ghiasi, Ali Shafahi, Reza Ardekani

We propose adaptive weight decay, which automatically tunes the hyper-parameter for weight decay during each training iteration. For classification problems, we propose changing the value of the weight decay hyper-parameter on the fly based on the strength of updates from the classification loss (i.e., gradient of cross-entropy), and the regularization loss

Raviteja Vemulapalli, Warren Richard Morningstar, Philip Andrew Mansfield, Hubert Eichner

Dual encoding models that encode a pair of inputs are widely used for representation learning. Many approaches train dual encoding models by maximizing agreement between pairs of encodings on centralized training data. However, in many scenarios, datasets are inherently decentralized across many clients (user devices or organizations) due to privacy concerns

Peng Zhao, Anirban Bhattacharya, Debdeep Pati, Bani K. Mallick

Modern data science applications often involve complex relational data with dynamic structures. An abrupt change in such dynamic relational data is typically observed in systems that undergo regime changes due to interventions. In such a case, we consider a factorized fusion shrinkage model in which all decomposed factors are dynamically shrunk towards group

Richard Mandel, Alexander Ushakov

Consider the equation $q_1α^{x_1}+\dots+q_kα^{x_k} = q$, with constants $α\in \overline{\mathbb{Q}} \setminus \{0,1\}$, $q_1,\ldots,q_k,q\in\overline{\mathbb{Q}}$ and unknowns $x_1,\ldots,x_k$, referred to in this paper as an \emph{algebraic equation with exponents}. We prove that the problem to decide if a given equation has an integer solution is $\textbf{

Diego Dominici, Juan José Moreno Balcázar

In this paper we tackle the asymptotic behavior of a family of orthogonal polynomials with respect to a nonstandard inner product involving the forward operator Δ. Concretely, we treat the generalized Charlier weights in the framework of Δ--Sobolev orthogonality. We obtain an asymptotic expansion for this orthogonal polynomials where the falling factorial po

Lin Gui, Victor Veitch

Consider the problem of estimating the causal effect of some attribute of a text document; for example: what effect does writing a polite vs. rude email have on response time? To estimate a causal effect from observational data, we need to adjust for confounding aspects of the text that affect both the treatment and outcome -- e.g., the topic or writing leve

Sina Hazratpour, Emily Riehl

Consider a locally cartesian closed category with an object I and a class of trivial fibrations, which admit sections and are stable under pushforward and retract as arrows. Define the fibrations to be those maps whose Leibniz exponential with the generic point of I defines a trivial fibration. Then the fibrations are also closed under pushforward.

Kwangyoun Kim, Felix Wu, Yifan Peng, Jing Pan

Conformer, combining convolution and self-attention sequentially to capture both local and global information, has shown remarkable performance and is currently regarded as the state-of-the-art for automatic speech recognition (ASR). Several other studies have explored integrating convolution and self-attention but they have not managed to match Conformer&#3

Anupam Ray

Dark Matter (DM) is omnipresent in our universe. Despite its abundance, the microscopic identity of DM still remains a mystery. Primordial black holes (PBHs), possibly formed via gravitational collapse of large density perturbations in the early universe, are one of the earliest proposed and viable DM candidates. Recent studies indicate that PBHs can make up

Souvik Roy, Dibyendu Nandy

To understand the global-scale physical processes behind coronal mass ejection (CME)-driven geomagnetic storms and predict their intensity as a space weather forecasting measure, we develop an interplanetary CME flux rope-magnetosphere interaction module using 3D magnetohydrodynamics. The simulations adequately describe ICME-forced dynamics of the magnetosph

Julius von Rohrscheidt, Bastian Rieck

The manifold hypothesis, which assumes that data lies on or close to an unknown manifold of low intrinsic dimension, is a staple of modern machine learning research. However, recent work has shown that real-world data exhibits distinct non-manifold structures, i.e. singularities, that can lead to erroneous findings. Detecting such singularities is therefore

Naman Shah, Siddharth Srivastava

This paper addresses the problem of reliably and efficiently solving broad classes of long-horizon stochastic path planning problems. Starting with a vanilla RL formulation with a stochastic dynamics simulator and an occupancy matrix of the environment, our approach computes useful options with policies as well as high-level paths that compose the discovered

Zheng Cao, Raymond Guo, Caesar M. Tuguinay, Mark Pock

This paper presents a methodology for combining programming and mathematics to optimize elevator wait times. Based on simulated user data generated according to the canonical three-peak model of elevator traffic, we first develop a naive model from an intuitive understanding of the logic behind elevators. We take into consideration a general array of feature

Donghan Yu, Sheng Zhang, Patrick Ng, Henghui Zhu

Question answering over knowledge bases (KBs) aims to answer natural language questions with factual information such as entities and relations in KBs. Previous methods either generate logical forms that can be executed over KBs to obtain final answers or predict answers directly. Empirical results show that the former often produces more accurate answers, b

Pouya Bashivan, Adam Ibrahim, Amirozhan Dehghani, Yifei Ren

Model ensembles have long been used in machine learning to reduce the variance in individual model predictions, making them more robust to input perturbations. Pseudo-ensemble methods like dropout have also been commonly used in deep learning models to improve generalization. However, the application of these techniques to improve neural networks' robust

Julian Kranz, Shintaro Nishikawa

Building on Enders--Schemeitat--Tikuisis' classification, we show that a separable $C^*$-algebra $A$ with approximately inner flip in the UCT class is $K$-theoretically self-absorbing if and only if for every finite group $G$, the Bernoulli shift on $A^{\otimes G}$ is $KK^G$-equivalent to the trivial action. This in particular applies to UHF-algebras of

Yurii Khomskii, Hrafn Valtýr Oddsson

We present a novel treatment of set theory in a four-valued paraconsistent and paracomplete logic, i.e., a logic in which propositions can be both true and false, and neither true nor false. Our approach is a significant departure from previous research in paraconsistent set theory, which has almost exclusively been motivated by a desire to avoid Russell&#39

Mikael Kurula

Thirty years after the introduction of port-Hamiltonian systems, interest in this system class still remains high among systems and control researchers. Very recently, Jacob and Laasri obtained strong results on the solvability and well-posedness of time-varying linear port-Hamil\-to\-nian systems with boundary control and boundary observation. In this paper

Saeid Asgari Taghanaki, Aliasghar Khani, Fereshte Khani, Ali Gholami

A fundamental challenge of over-parameterized deep learning models is learning meaningful data representations that yield good performance on a downstream task without over-fitting spurious input features. This work proposes MaskTune, a masking strategy that prevents over-reliance on spurious (or a limited number of) features. MaskTune forces the trained mod

Reza Nasirigerdeh, Javad Torkzadehmahani, Daniel Rueckert, Georgios Kaissis

Normalization is an important but understudied challenge in privacy-related application domains such as federated learning (FL), differential privacy (DP), and differentially private federated learning (DP-FL). While the unsuitability of batch normalization for these domains has already been shown, the impact of other normalization methods on the performance

Danyu Zhang

We construct non-algebraic Anosov flows in dimension $3+2n$, $n\geq 2$, by suspending the action of the fundamental group of a finite cover of the Bonatti-Langevin flow.

Jeremy A. Collins, Patrick Grady, Charles C. Kemp

Robotic manipulation can benefit from wrist-mounted force/torque (F/T) sensors, but conventional F/T sensors can be expensive, difficult to install, and damaged by high loads. We present Visual Force/Torque Sensing (VFTS), a method that visually estimates the 6-axis F/T measurement that would be reported by a conventional F/T sensor. In contrast to approache

Venkatraman Renganathan, Joshua Pilipovsky, Panagiotis Tsiotras

This article extends the optimal covariance steering (CS) problem for discrete time linear stochastic systems modeled using moment-based ambiguity sets. To hedge against the uncertainty in the state distributions while performing covariance steering, distributionally robust risk constraints are employed during the optimal allocation of the risk. Specifically

Jan Christoph Schlegel, Mateusz Kwaśnicki, Akaki Mamageishvili

We study axiomatic foundations for different classes of constant-function automated market makers (CFMMs). We focus particularly on separability and on different invariance properties under scaling. Our main results are an axiomatic characterization of a natural generalization of constant product market makers (CPMMs), popular in decentralized finance, on th

Kim Klinger-Logan, Stephen D. Miller, Danylo Radchenko

We complete the program, initiated in a 2015 paper of Green, Miller, and Vanhove, of directly constructing the automorphic solution to the string theory $D^6 R^4$ differential equation $(Δ-12)f=-E_{3/2}^2$ for $SL(2,\Z)$. The construction is via a type of Poincaré series, and requires explicitly evaluating a particular double integral. We also show how to us

Mavina Nikandrou, Lu Yu, Alessandro Suglia, Ioannis Konstas

Continual learning aims to train a model incrementally on a sequence of tasks without forgetting previous knowledge. Although continual learning has been widely studied in computer vision, its application to Vision+Language tasks is not that straightforward, as settings can be parameterized in multiple ways according to their input modalities. In this paper,

CMS Collaboration

A search for new heavy resonances decaying to WW, WZ, ZZ, WH, or ZH boson pairs in the all-jets final state is presented. The analysis is based on proton-proton collision data recorded by the CMS detector in 2016-2018 at a centre-of-mass energy of 13 TeV at the CERN LHC, corresponding to an integrated luminosity of 138 fb$^{-1}$. The search is sensitive to r

Babatunji Omoniwa, Boris Galkin, Ivana Dusparic

Unmanned aerial vehicles (UAVs) are increasingly deployed to provide wireless connectivity to static and mobile ground users in situations of increased network demand or points of failure in existing terrestrial cellular infrastructure. However, UAVs are energy-constrained and experience the challenge of interference from nearby UAV cells sharing the same fr

Bart Kieboom, Maria Bartzioka, Matin Jafarian

This paper studies the problem of output regulation for a class of nonlinear systems experiencing matched input disturbances. It is assumed that the disturbance signal is generated by an external autonomous dynamical system. First, we show that for a class of nonlinear systems admitting a finite-dimensional Koopman representation, the problem is equivalent t

Zhiqi Bu, Yu-Xiang Wang, Sheng Zha, George Karypis

Differentially private (DP) optimization is the standard paradigm to learn large neural networks that are accurate and privacy-preserving. The computational cost for DP deep learning, however, is notoriously heavy due to the per-sample gradient clipping. Existing DP implementations are 2-1000X more costly in time and space complexity than the standard (non-p

Koresh Khateri, Karthik Soma, Mahdi Pourgholi, Mohsen Montazeri

The topology of a robotic swarm affects the convergence speed of consensus and the mobility of the robots. In this paper, we prove the existence of a complete set of local topology manipulation operations that allow the transformation of a swarm topology. The set is complete in the sense that any other possible set of manipulation operations can be performed

Leandro von Werra, Lewis Tunstall, Abhishek Thakur, Alexandra Sasha Luccioni

Evaluation is a key part of machine learning (ML), yet there is a lack of support and tooling to enable its informed and systematic practice. We introduce Evaluate and Evaluation on the Hub --a set of tools to facilitate the evaluation of models and datasets in ML. Evaluate is a library to support best practices for measurements, metrics, and comparisons of

Zhiqi Bu, Yu-Xiang Wang, Sheng Zha, George Karypis

We study the problem of differentially private (DP) fine-tuning of large pre-trained models -- a recent privacy-preserving approach suitable for solving downstream tasks with sensitive data. Existing work has demonstrated that high accuracy is possible under strong privacy constraint, yet requires significant computational overhead or modifications to the ne

Goetz Graefe, Thanh Do

Recent work shows how offset-value coding speeds up database query execution, not only sorting but also duplicate removal and grouping (aggregation) in sorted streams, order-preserving exchange (shuffle), merge join, and more. It already saves thousands of CPUs in Google's Napa and F1 Query systems, e.g., in grouping algorithms and in log-structured merg

Lilia Anguelova, Calin Iuliu Lazaroiu

We derive consistency conditions for sustained slow roll and rapid turn inflation in two-field cosmological models with oriented scalar field space, which imply that inflationary models with field-space trajectories of this type are non-generic. In particular, we show that third order adiabatic slow roll, together with large and slowly varying turn rate, req

Yecheng Jason Ma, Shagun Sodhani, Dinesh Jayaraman, Osbert Bastani

Reward and representation learning are two long-standing challenges for learning an expanding set of robot manipulation skills from sensory observations. Given the inherent cost and scarcity of in-domain, task-specific robot data, learning from large, diverse, offline human videos has emerged as a promising path towards acquiring a generally useful visual re

Harlan Campbell, Paul Gustafson

In many common situations, a Bayesian credible interval will be, given the same data, very similar to a frequentist confidence interval, and researchers will interpret these intervals in a similar fashion. However, no predictable similarity exists when credible intervals are based on model-averaged posteriors whenever one of the two nested models under consi

Rodrigo Bañuelos, Mateusz Kwaśnicki

The long-standing conjecture that for $p \in (1, \infty)$ the $\ell^p(\mathbb Z)$ norm of the Riesz--Titchmarsh discrete Hilbert transform is the same as the $L^p(\mathbb R)$ norm of the classical Hilbert transform, is verified when $p = 2 n$ or $\frac{p}{p - 1} = 2 n$, for $n \in \mathbb N$. The proof, which is algebraic in nature, depends in a crucial way

Siddhartha Banerjee, Sean R. Sinclair, Milind Tambe, Lily Xu

Most real-world deployments of bandit algorithms exist somewhere in between the offline and online set-up, where some historical data is available upfront and additional data is collected dynamically online. How best to incorporate historical data to "warm start" bandit algorithms is an open question: naively initializing reward estimates using all h

Henning Bahl, Marcela Carena, Nina M. Coyle, Aurora Ireland

Two Higgs doublet models (2HDM) provide the low energy effective field theory (EFT) description in many well motivated extensions of the Standard Model. It is therefore relevant to study their properties, as well as the theoretical constraints on these models. In this article we concentrate on three relevant requirements for the validity of the 2HDM framewor

Yue Yu

On the Sb-layer of the Kagome superconductor CsV$_3$Sb$_5$, pair density wave states have been observed. When the high-temperature charge orderings are treated as static backgrounds, these PDW states exhibit the same wavevector in the effective 2D Brillouin zone. Interestingly, these PDW states break the same symmetry on the surface. Considering the presence

Sin Kyu Kang, Jinheung Kim, Soojin Lee, Jeonghyeon Song

The type-I two-Higgs-doublet model in the inverted Higgs scenario can retain the theoretical stability all the way up to the Planck scale. The Planck-cutoff scale, $Λ_{\rm cut}^{\rm Planck}$, directly impacts the mass spectra such that all the extra Higgs boson masses should be light below about 160 GeV. However, the observation of the light masses of new Hi

Anja Butter, Theo Heimel, Till Martini, Sascha Peitzsch

The matrix element method is widely considered the ultimate LHC inference tool for small event numbers. We show how a combination of two conditional generative neural networks encodes the QCD radiation and detector effects without any simplifying assumptions, while keeping the computation of likelihoods for individual events numerically efficient. We illustr

Alex May, Jonathan Sorce, Beni Yoshida

In the AdS/CFT correspondence, bulk causal structure has consequences for boundary entanglement. In quantum information science, causal structures can be replaced by distributed entanglement for the purposes of information processing. In this work, we deepen the understanding of both of these statements, and their relationship, with a number of new results.

Alexander J. Millar, Steven M. Anlage, Rustam Balafendiev, Pavel Belov

We summarise the recent progress of the Axion Longitudinal Plasma HAloscope (ALPHA) Consortium, a new experimental collaboration to build a plasma haloscope to search for axions and dark photons. The plasma haloscope is a novel method for the detection of the resonant conversion of light dark matter to photons. ALPHA will be sensitive to QCD axions over almo

Patrycja Łydżba, Marcin Mierzejewski, Marcos Rigol, Lev Vidmar

Thermalization (generalized thermalization) in nonintegrable (integrable) quantum systems requires two ingredients: equilibration and agreement with the predictions of the Gibbs (generalized Gibbs) ensemble. We prove that observables that exhibit eigenstate thermalization in single-particle sector equilibrate in many-body sectors of quantum-chaotic quadratic

Nicole M. Khusid, Chiara M. F. Mingarelli, Priyamvada Natarajan, J. Andrew Casey-Clyde

Supermassive black hole binary systems (SMBHBs) should be the most powerful sources of gravitational waves (GWs) in the Universe. Once Pulsar Timing Arrays (PTAs) detect the stochastic GW background from their cosmic merger history, searching for individually resolvable binaries will take on new importance. Since these individual SMBHBs are expected to be ra

Ricardo Yarza, Rosa Wallace Everson, Enrico Ramirez-Ruiz

Common-envelope evolution is a stage in binary system evolution in which a giant star engulfs a companion. The standard energy formalism is an analytical framework to estimate the amount of energy transferred from the companion's shrinking orbit into the envelope of the star that engulfed it. We show analytically that this energy transfer is larger than

Ananthan Karunakaran, Dennis Zaritsky

We extend the Ultra-Diffuse Galaxy (UDG) abundance relation, $N_{UDG}-M_{200}$, to lower halo mass hosts $(M_{200}\sim10^{11.6-12.2}M_{\odot})$. We select UDG satellites from published catalogs of dwarf satellite galaxies around Milky Way analogs, namely the Exploration of Local Volume Satellites (ELVES) survey, Satellite Around Galactic Analogs (SAGA) surve

Weicheng Kuo, Yin Cui, Xiuye Gu, AJ Piergiovanni

We present F-VLM, a simple open-vocabulary object detection method built upon Frozen Vision and Language Models. F-VLM simplifies the current multi-stage training pipeline by eliminating the need for knowledge distillation or detection-tailored pretraining. Surprisingly, we observe that a frozen VLM: 1) retains the locality-sensitive features necessary for d

Emilio H. S. Sousa, A. Vidiella-Barranco, J. A. Roversi

We investigate the generation and transfer of entangled states between two coupled microtoroidal cavities considering two different types of couplings, namely i) via a bridge qubit and ii) via evanescent fields. The cavities support two counter-propagating whispering-gallery modes (WGMs) that may also interact with each other. We firstly show that it is poss

Wenjie Li, Shu-Tao Xia, Jiangke Fan, Teng Zhang

Traditional vertical federated learning schema suffers from two main issues: 1) restricted applicable scope to overlapped samples and 2) high system challenge of real-time federated serving, which limits its application to advertising systems. To this end, we advocate a new practical learning setting, Semi-VFL (Vertical Semi-Federated Learning), for real-wor

Alexandre Belsley

Absorption spectroscopy is a widely used technique that permits the detection and characterization of gas species at low concentrations. We propose a sensing strategy combining the advantages of frequency modulation spectroscopy with the reduced noise properties accessible by squeezing the probe state. A homodyne detection scheme allows the simultaneous meas

Dong Xiao, Zuoqiang Shi, Siyu Li, Bailin Deng

Oriented normals are common pre-requisites for many geometric algorithms based on point clouds, such as Poisson surface reconstruction. However, it is not trivial to obtain a consistent orientation. In this work, we bridge orientation and reconstruction in the implicit space and propose a novel approach to orient point cloud normals by incorporating isovalue

Jordi Grau-Moya, Grégoire Delétang, Markus Kunesch, Tim Genewein

Meta-training agents with memory has been shown to culminate in Bayes-optimal agents, which casts Bayes-optimality as the implicit solution to a numerical optimization problem rather than an explicit modeling assumption. Bayes-optimal agents are risk-neutral, since they solely attune to the expected return, and ambiguity-neutral, since they act in new situat

G. E. Volovik

Diakonov theory of quantum gravity, in which tetrads emerge as the bilinear combinations of the fermionis fields,\cite{Diakonov2011} suggests that in general relativity the metric may have dimension 2, i.e. $[g_{μν}]=1/[L]^2$. Several other approaches to quantum gravity, including the model of superplastic vacuum and $BF$-theories of gravity support this sug

Timothy Greco, Daniel E. Koditschek

This technical report provides a more thorough treatment of the proofs and derivations contained in a recent conference paper submission. The description of the anchoring controller is reproduced here without abridgement, and additional appendices provide a clearer account of the implementation details.

Jayesh K. Gupta, Johannes Brandstetter

Partial differential equations (PDEs) are central to describing complex physical system simulations. Their expensive solution techniques have led to an increased interest in deep neural network based surrogates. However, the practical utility of training such surrogates is contingent on their ability to model complex multi-scale spatio-temporal phenomena. Va

Omid Amini, Lucas Gierczak

We develop a purely combinatorial theory of limit linear series on metric graphs. This will be based on the formalisms of hypercube rank functions and slope structures. We provide a full classification of combinatorial limit linear series of rank one, and discuss connections to other concepts in tropical algebra and combinatorial algebraic geometry.

Kevin E. Wu, Kevin K. Yang, Rianne van den Berg, James Y. Zou

The ability to computationally generate novel yet physically foldable protein structures could lead to new biological discoveries and new treatments targeting yet incurable diseases. Despite recent advances in protein structure prediction, directly generating diverse, novel protein structures from neural networks remains difficult. In this work, we present a

Annegret Burtscher, Leonardo García-Heveling

No Hopf-Rinow Theorem is possible in Lorentzian Geometry. Nonetheless, we prove that a spacetime is globally hyperbolic if and only if it is metrically complete with respect to the null distance of a time function. Our approach is based on the observation that null distances behave particularly well for weak temporal functions in terms of regularity and caus

Alex Glyn-Davies, Connor Duffin, Ö. Deniz Akyildiz, Mark Girolami

Incorporating unstructured data into physical models is a challenging problem that is emerging in data assimilation. Traditional approaches focus on well-defined observation operators whose functional forms are typically assumed to be known. This prevents these methods from achieving a consistent model-data synthesis in configurations where the mapping from

Philippe Moreillon

We consider the free additive convolution semigroup $\lbrace μ^{\boxplus t}:\,t\ge 1\rbrace$ and determine the local behavior of the density of $μ^{\boxplus t}$ at the endpoints and at any singular point of its support. We then study the free additive convolution of two multi-cut probability measures and show that its density decays either as a square root o

Harshit Yadav

Let $U:\mathcal{C}\rightarrow\mathcal{D}$ be a strong monoidal functor between abelian monoidal categories admitting a right adjoint $R$, such that $R$ is exact, faithful and the adjunction $U\dashv R$ is coHopf. Building on the work of Balan, we show that $R$ is separable (resp., special) Frobenius monoidal if and only if $R(\mathbb{1}_{\mathcal{D}})$ is a

Maan Qraitem, Kate Saenko, Bryan A. Plummer

Prior work has shown that Visual Recognition datasets frequently underrepresent bias groups $B$ (\eg Female) within class labels $Y$ (\eg Programmers). This dataset bias can lead to models that learn spurious correlations between class labels and bias groups such as age, gender, or race. Most recent methods that address this problem require significant archi

Šárka Todorova-Nová

A brief overview of properties of the model of quantized fragmentation of a helical QCD string is followed by the discussion of colour reconnection models and of the ways to reproduce the evolution of the average transverse momentum in hadronic events, as function of charged particle multiplicity. Reference to recent experimental data is included.

Nafis Irtija, Jim Plusquellic, Eirini Eleni Tsiropoulou, Joshua Goldberg

Electronic control systems used for quantum computing have become increasingly complex as multiple qubit technologies employ larger numbers of qubits with higher fidelity targets. Whereas the control systems for different technologies share some similarities, parameters like pulse duration, throughput, real-time feedback, and latency requirements vary widely

James Davies

We prove that every finite colouring of the plane contains a monochromatic pair of points at an odd distance from each other.

Hung Nghiep Tran, Atsuhiro Takasu

Knowledge graph embedding aims to predict the missing relations between entities in knowledge graphs. Tensor-decomposition-based models, such as ComplEx, provide a good trade-off between efficiency and expressiveness, that is crucial because of the large size of real world knowledge graphs. The recent multi-partition embedding interaction (MEI) model subsume

Antti Koskela, Marlon Tobaben, Antti Honkela

Individual privacy accounting enables bounding differential privacy (DP) loss individually for each participant involved in the analysis. This can be informative as often the individual privacy losses are considerably smaller than those indicated by the DP bounds that are based on considering worst-case bounds at each data access. In order to account for the

Alex Damian, Eshaan Nichani, Jason D. Lee

Traditional analyses of gradient descent show that when the largest eigenvalue of the Hessian, also known as the sharpness $S(θ)$, is bounded by $2/η$, training is "stable" and the training loss decreases monotonically. Recent works, however, have observed that this assumption does not hold when training modern neural networks with full batch or larg

Nour-Eddine Abouelkhir, Hanane EL Hadfi, Abdallah Slaoui, Rachid Ahl Laamara

In statistical estimation theory, it has been shown previously that the Wigner-Yanase skew information is bounded by the quantum Fisher information associated with the phase parameter. Besides, the quantum Cramér-Rao inequality is expressed in terms of skew information. Since these two fundamental quantities are based on the concept of quantum uncertainty, w

Wen-Qing Guo, Yichao Li, Xiaoyuan Huang, Yin-Zhe Ma

The Galactic center $γ$-ray excess, detected by the Fermi-LAT, is a very attractive tentative signal from dark matter annihilation. Searching for associated synchrotron emissions can test the dark matter interpretation for this excess. We geared the Five-hundred-meter Aperture Spherical radio Telescope (FAST) towards Coma Berenices (a dwarf Spheroidal galaxy

Skanda Koppula, Yazhe Li, Evan Shelhamer, Andrew Jaegle

Self-supervised methods have achieved remarkable success in transfer learning, often achieving the same or better accuracy than supervised pre-training. Most prior work has done so by increasing pre-training computation by adding complex data augmentation, multiple views, or lengthy training schedules. In this work, we investigate a related, but orthogonal q

Vladimir Mitankin, Masahiro Nakahara, Sam Streeter

We study local-global principles for two notions of semi-integral points, termed Campana points and Darmon points. In particular, we develop a semi-integral version of the Brauer-Manin obstruction interpolating between Manin's classical version for rational points and the integral version developed by Colliot-Thélène and Xu. We determine the status of lo

Diksha Garg, Sameer Patel, Mary Hall Reno, Alexander Reustle

Ultra-high-energy neutrinos serve as messengers of some of the highest energy astrophysical environments. Given that neutrinos are neutral and only interact via weak interactions, neutrinos can emerge from sources, traverse astronomical distances, and point back to their origins. Their weak interactions require large target volumes for neutrino detection. Us

Kiarash Jamali, Dari Kimanius, Sjors H. W. Scheres

Electron cryo-microscopy (cryo-EM) produces three-dimensional (3D) maps of the electrostatic potential of biological macromolecules, including proteins. Along with knowledge about the imaged molecules, cryo-EM maps allow de novo atomic modelling, which is typically done through a laborious manual process. Taking inspiration from recent advances in machine le

Hans U. Boden, Ceyhun Elmacioglu, Anshul Guha, Homayun Karimi

We define a knot to be half ribbon if it is the cross-section of a ribbon 2-knot, and observe that ribbon implies half ribbon implies slice. We introduce the half ribbon genus of a knot K, the minimum genus of a ribbon knotted surface of which K is a cross-section. We compute this genus for all prime knots up to 12 crossings, and many 13-crossing knots. The

Yan Gao, Javier Fernandez-Marques, Titouan Parcollet, Pedro P. B. de Gusmao

Self-supervised learning (SSL) has proven vital in speech and audio-related applications. The paradigm trains a general model on unlabeled data that can later be used to solve specific downstream tasks. This type of model is costly to train as it requires manipulating long input sequences that can only be handled by powerful centralised servers. Surprisingly

Benjamin Russo, M. Paul Laiu

In this paper, we give an in-depth error analysis for surrogate models generated by a variant of the Sparse Identification of Nonlinear Dynamics (SINDy) method. We start with an overview of a variety of non-linear system identification techniques, namely, SINDy, weak-SINDy, and the occupation kernel method. Under the assumption that the dynamics are a finite

Mar Bastero-Gil, António Torres Manso

Inflaton-vector interactions of the type $ϕF\tilde{F}$ have provided interesting phenomenology to tackle some of current problems in cosmology, namely the vectors could constitute the dark matter component. It could also lead to possible signatures imprinted in a gravitational wave spectrum. Through this coupling, a rolling inflaton induces an exponential pr