May 2022 arXiv papers — page 29

Timmy S. T. Wan, Jun-Cheng Chen, Tzer-Yi Wu, Chu-Song Chen

In visual search, the gallery set could be incrementally growing and added to the database in practice. However, existing methods rely on the model trained on the entire dataset, ignoring the continual updating of the model. Besides, as the model updates, the new model must re-extract features for the entire gallery set to maintain compatible feature space,

Zhenting Wang, Juan Zhai, Shiqing Ma

Deep neural networks are vulnerable to Trojan attacks. Existing attacks use visible patterns (e.g., a patch or image transformations) as triggers, which are vulnerable to human inspection. In this paper, we propose stealthy and efficient Trojan attacks, BppAttack. Based on existing biology literature on human visual systems, we propose to use image quantizat

Xinwei Li, Junichiro Kono, Qimiao Si, Silke Paschen

Strange metal behavior appears across a variety of condensed matter settings and beyond, and achieving a universal understanding is an exciting prospect. The beyond-Landau quantum criticality of Kondo destruction has had considerable success in describing the behavior of strange metal heavy fermion compounds, and there is some evidence that the associated pa

Miguel Pereira

We establish computational results concerning the Lagrangian capacity from "Cieliebak and Mohnke - Punctured holomorphic curves and Lagrangian embeddings". More precisely, we show that the Lagrangian capacity of a 4-dimensional convex toric domain is equal to its diagonal. The proof involves comparisons between the Lagrangian capacity, the McDuff-Siegel capa

Amanda Fernández-Fontelo, Felix Henninger, Pascal J. Kieslich, Frauke Kreuter

We propose new ensemble models for multivariate functional data classification as combinations of semi-metric-based weak learners. Our models extend current semi-metric-type methods from the univariate to the multivariate case, propose new semi-metrics to compute distances between functions, and consider more flexible options for combining weak learners usin

Edwin Eobaldt, Francesco Vitale, Maximilian Zapf, Margarita Lapteva

Mixed dimensional hybrid structures have recently gained increasing attention as promising building blocks for novel electronic and optoelectronic devices. In this context, hybridization of semiconductor nanowires with two-dimensional materials could offer new ways to control and modulate lasing at the nanoscale. In this work, we deterministically fabricate

Matias Risaro, Paolo Savio, Marco Pizzocaro, Filippo Levi

The advent of optical frequency standards with ultimate uncertainties in the low 1x10$^{-18}$ requires femtosecond frequency combs to support a similar level of resolution in the spectral transfer and the computation of optical frequency ratios. The related experimental challenges grow together with the number of optical frequencies to be measured simultaneo

Martina Hofmanova, Rongchan Zhu, Xiangchan Zhu

We study the surface quasi-geostrophic equation with an irregular spatial perturbation $$ \partial_{t }\theta+ u\cdot\nabla\theta = -\nu(-\Delta)^{\gamma/2}\theta+ \zeta,\qquad u=\nabla^{\perp}(-\Delta)^{-1}\theta, $$ on $[0,\infty)\times\mathbb{T}^{2}$, with $\nu\geq 0$, $\gamma\in [0,3/2)$ and $\zeta\in B^{-2+\kappa}_{\infty,\infty}(\mathbb{T}^{2})$ for so

Hillel Sanhedrai, Shlomo Havlin

Multi-stability is a widely observed phenomenon in real complex networked systems, such as technological infrastructures, ecological systems, gene regulation, transportation and more. When a system functions normally but there exists also a potential state with abnormal low activity, although the system is at equilibrium it might make a transition into the l

Zi-Qi Lian, You-Yang Zhou, Liu-Jun Wang, Qing Chen

In quantum information, it is of high importance to efficiently detect entanglement. Generally, it needs quantum tomography to obtain state density matrix. However, it would consumes a lot of measurement resources, and the key is how to reduce the consumption. In this paper, we discovered the relationship between convolutional layer of artificial neural netw

Hiroyuki Nakagawa, Shinya Tsuchida, Emiliano Tramontana, Andrea Fornaia

Embedded systems including IoT devices are designed for specialized functions; thus, changes in functions are not considered following their release. For this reason, changing functions to satisfy the requirements of IoT systems is difficult. In this study, we focus on updating existing embedded systems without modifying them. We investigate the design of ne

Mrityunjoy Ghosh, Sheela Verma

In this paper, we study the shape optimization problem for the first eigenvalue of the $p$-Laplace operator with the mixed Neumann-Dirichlet boundary conditions on multiply-connected domains in hyperbolic space. Precisely, we establish that among all multiply-connected domains of a given volume and prescribed $(n-1)$-th quermassintegral of the convex Dirichl

Sarah Isufi, Kristijan Poje, Igor Vukobratovic, Mario Brcic

We shall have a hard look at ethics and try to extract insights in the form of abstract properties that might become tools. We want to connect ethics to games, talk about the performance of ethics, introduce curiosity into the interplay between competing and coordinating in well-performing ethics, and offer a view of possible developments that could unify in

Prabal Adhikari, Elizabeth Leeser, Jake Markowski

We study phonon modes associated with magnetic vortex lattices of finite isospin chiral perturbation theory near the upper critical point by introducing quasimomentum fluctuations to the lattice and calculate dispersion relations associated with the optical and acoustic modes. We find that one of the acoustic modes is massless and that its energy for small t

Wonjun Kang, Geonsu Lee, Hyung Il Koo, Nam Ik Cho

The goal of face reenactment is to transfer a target expression and head pose to a source face while preserving the source identity. With the popularity of face-related applications, there has been much research on this topic. However, the results of existing methods are still limited to low-resolution and lack photorealism. In this work, we present a one-sh

Ohad Silbiger, Yakir Hadad

The inverse-scattering problem of an inhomogeneous material has been of interest for many years, and was generally addressed with various optimization techniques. In this paper, we suggest an optimization-free method for solving the inverse-scattering problem of a one-dimensional inhomogeneous medium and use this to demonstrate the design of desired reflecti

Zdzislaw Brzezniak, Benedetta Ferrario, Margherita Zanella

We study the nonlinear Schr\"odinger equation with linear damping, i.e. a zero order dissipation, and additive noise. Working in $R^d$ with d = 2 or d = 3, we prove the uniqueness of the invariant measure when the damping coefficient is sufficiently large.

Sam Yang, Bjorn Vaagensmith, Deepika Patra, Ryan Hruska

We propose a multi-fidelity neural network (MFNN) tailored for rapid high-dimensional grid power flow simulations and contingency analysis with scarce high-fidelity contingency data. The proposed model comprises two networks -- the first one trained on DC approximation as low-fidelity data and coupled to a high-fidelity neural net trained on both low- and hi

Ahmad Salmanogli

In this study, a circuitry model of the coupling of a qubit to reservoir modes is defined to clearly determine the effect of the reservoir modes on the qubit decay and dephasing rates. The main goal is to theoretically calculate the dephasing and decay rate of a qubit, particularly due to the circuitry effect. Firstly, the Hamiltonian of the system (coupling

Alexander F. Goncharov, Vitali B. Prakapenka

Weck et al. (1) report on the existence and stability fields of two superionic (SI) phases of H2O ice at high P-T (P-T) conditions, which has been a topic of static and dynamic experiments and theoretical calculations (see Ref. (2) and references therein). They confirm Ref. (2) in that there are two SI phases with bcc and fcc oxygen sublattices with the stab

Xiao Zhang, Dejing Dou, Ji Wu

In continual and lifelong learning, good representation learning can help increase performance and reduce sample complexity when learning new tasks. There is evidence that representations do not suffer from "catastrophic forgetting" even in plain continual learning, but little further fact is known about its characteristics. In this paper, we aim to gain mor

Tong Wei, Qian-Yu Liu, Jiang-Xin Shi, Wei-Wei Tu

Long-Tailed Semi-Supervised Learning (LTSSL) aims to learn from class-imbalanced data where only a few samples are annotated. Existing solutions typically require substantial cost to solve complex optimization problems, or class-balanced undersampling which can result in information loss. In this paper, we present the TRAS (TRAnsfer and Share) to effectively

Zhang Bingyu, Nikolay Arefyev

The current state-of-the-art test accuracy (97.42\%) on the IMDB movie reviews dataset was reported by \citet{thongtan-phienthrakul-2019-sentiment} and achieved by the logistic regression classifier trained on the Document Vectors using Cosine Similarity (DV-ngrams-cosine) proposed in their paper and the Bag-of-N-grams (BON) vectors scaled by Naive Bayesian

Sira Gratz, Greg Stevenson

We initiate a systematic study of lattices of thick subcategories for arbitrary essentially small triangulated categories. To this end we give several examples illustrating the various properties these lattices may, or may not, have and show that as soon as a lattice of thick subcategories is distributive it is automatically a spatial frame. We then construc

Erin Carson, Ieva Daužickaitė

Low rank matrix approximations appear in a number of scientific computing applications. We consider the Nystr\"{o}m method for approximating a positive semidefinite matrix $A$. In the case that $A$ is very large or its entries can only be accessed once, a single-pass version may be necessary. In this work, we perform a complete rounding error analysis of the

Jian-Bo Cheng, Zi-Yang Lin, Shi-Lin Zhu

The LHCb Collaboration discovered a double-charm tetraquark $T_{cc}^{+}$ with a very small width. We investigate the $T_{cc}^{+}$ as a $DD^{*}$ molecule with $J^{P}=1^{+}$ in the framework of the one-boson-exchange potential model. The isospin breaking effect and $S-D$ wave coupling are taken into account carefully. We adopt the complex scaling method (CSM)

Arian Askari, Georgios Peikos, Gabriella Pasi, Suzan Verberne

This paper summarizes our approaches submitted to the case law retrieval task in the Competition on Legal Information Extraction/Entailment (COLIEE) 2022. Our methodology consists of four steps; in detail, given a legal case as a query, we reformulate it by extracting various meaningful sentences or n-grams. Then, we utilize the pre-processed query case to r

Daniele Boffi, Fabio Credali, Lucia Gastaldi

We study a recent formulation for fluid-structure interaction problems based on the use of a distributed Lagrange multiplier in the spirit of the fictitious domain approach. In this paper, we focus our attention on a crucial computational aspect regarding the interface matrix for the finite element discretization: it involves integration of functions support

Manuel Traub, Sebastian Otte, Tobias Menge, Matthias Karlbauer

Our brain can almost effortlessly decompose visual data streams into background and salient objects. Moreover, it can anticipate object motion and interactions, which are crucial abilities for conceptual planning and reasoning. Recent object reasoning datasets, such as CATER, have revealed fundamental shortcomings of current vision-based AI systems, particul

Tania Paul, V. Fernández Becerra, Timo Hyart

It has been proposed that band-inverted electron-hole bilayers support a phase transition from an insulating phase with spontaneously broken time-reversal symmetry to a quantum spin Hall insulator phase as a function of increasing electron and hole densities. Here, we show that in the presence of proximity-induced superconductivity it is possible to realize

Mingzhe Li, XieXiong Lin, Xiuying Chen, Jinxiong Chang

Contrastive learning has achieved impressive success in generation tasks to militate the "exposure bias" problem and discriminatively exploit the different quality of references. Existing works mostly focus on contrastive learning on the instance-level without discriminating the contribution of each word, while keywords are the gist of the text and dominant

Jianzong Wang, Shijing Si, Zhitao Zhu, Xiaoyang Qu

Deep learning models have made significant progress in automatic program repair. However, the black-box nature of these methods has restricted their practical applications. To address this challenge, this paper presents an interpretable approach for program repair based on sequence-to-sequence models with causal inference and our method is called CPR, short

Ran Ben Basat, Shay Vargaftik, Amit Portnoy, Gil Einziger

Distributed Mean Estimation (DME), in which $n$ clients communicate vectors to a parameter server that estimates their average, is a fundamental building block in communication-efficient federated learning. In this paper, we improve on previous DME techniques that achieve the optimal $O(1/n)$ Normalized Mean Squared Error (NMSE) guarantee by asymptotically i

Xingjian Li, Pengkun Yang, Yangcheng Gu, Xueying Zhan

Uncertainty estimation for unlabeled data is crucial to active learning. With a deep neural network employed as the backbone model, the data selection process is highly challenging due to the potential over-confidence of the model inference. Existing methods resort to special learning fashions (e.g. adversarial) or auxiliary models to address this challenge.

Xiuying Chen, Hind Alamro, Mingzhe Li, Shen Gao

The related work section is an important component of a scientific paper, which highlights the contribution of the target paper in the context of the reference papers. Authors can save their time and effort by using the automatically generated related work section as a draft to complete the final related work. Most of the existing related work section genera

S. Mahesh Chandran, S. Shankaranarayanan

In time-independent quantum systems, entanglement entropy possesses an inherent scaling symmetry that the energy of the system does not have. The symmetry also assures that entropy divergence can be associated with the zero modes. We generalize this symmetry to time-dependent systems all the way from a coupled harmonic oscillator with a time-dependent freque

Habib Ammari, Bryn Davies, Erik Orvehed Hiltunen

In this paper, we use recent breakthroughs in the study of coupled subwavelength resonator systems to reveal new insight into the mechanisms responsible for the fundamental features of Anderson localization. The occurrence strong localization in random media has proved difficult to understand, particularly in physically derived multi-dimensional models and s

Egor Voronetsky

We give an explicit description of internal actions in the semi-abelian categories of pro-groups and non-unital pro-rings in terms of actions of group objects and ring objects in $\mathrm{Pro}(\mathbf{Set})$, as well as in some related categories. Also, we show that a similar result fails for Lie algebras.

Chu Xu, Fen Tang, Qingqing Shang, Yao Fan

Photonic hooks (PHs) are non-evanescent light beams with a highly concentrated curved optical fields. Since their discovery, PHs always have one single inflection point and thus have a hook-like structure. In this work, a new type of PHs with two inflection points and S-shaped structures (S-PHs) were reported for the first time. We theoretically studied the

Ben S. Lakeland, Tim Naylor

Using simultaneous optical and infrared light curves of disc-bearing young stars in NGC 2264, we perform the first multi-wavelength structure function study of YSOs. We find that dippers have larger variability amplitudes than bursters and symmetric variables at all timescales longer than a few hours. By analysing optical-infrared colour time-series, we also

Mayank Raikwar, Danilo Gligoroski

The scientific interest in the area of Decentralized Randomness Beacon (DRB) protocols has been thriving recently. Partially that interest is due to the success of the disruptive technologies introduced by modern cryptography, such as cryptocurrencies, blockchain technologies, and decentralized finances, where there is an enormous need for a public, reliable

M. Herrero-Valea, A. S. Koshelev, A. Tokareva

Scattering amplitudes mediated by graviton exchange display IR singularities in the forward limit. This obstructs standard application of positivity bounds based on twice subtracted dispersion relations. Such divergences can be cancelled only if the UV limit of the scattering amplitude behaves in a specific way, which implies a very non-trivial connection be

Omer Belhasin, Guy Bar-Shalom, Ran El-Yaniv

This paper deals with deep transductive learning, and proposes TransBoost as a procedure for fine-tuning any deep neural model to improve its performance on any (unlabeled) test set provided at training time. TransBoost is inspired by a large margin principle and is efficient and simple to use. Our method significantly improves the ImageNet classification pe

MohammadTaghi Hajiaghayi, Max Springer

In this paper, we analyze a natural learning algorithm for uniform pacing of advertising budgets, equipped to adapt to varying ad sale platform conditions. On the demand side, advertisers face a fundamental technical challenge in automating bidding in a way that spreads their allotted budget across a given campaign subject to hidden, and potentially dynamic,

Begüm Ateşli, Oğul Esen, Manuel de León, Cristina Sardón

Cosymplectic geometry has been proven to be a very useful geometric background to describe time-dependent Hamiltonian dynamics. In this work, we address the globalization problem of locally cosymplectic Hamiltonian dynamics that failed to be globally defined. We investigate both the geometry of locally conformally cosymplectic (abbreviated as LCC) manifolds

Jiarui Feng, Yixin Chen, Fuhai Li, Anindya Sarkar

The most popular design paradigm for Graph Neural Networks (GNNs) is 1-hop message passing -- aggregating information from 1-hop neighbors repeatedly. However, the expressive power of 1-hop message passing is bounded by the Weisfeiler-Lehman (1-WL) test. Recently, researchers extended 1-hop message passing to K-hop message passing by aggregating information

Zhe-Qi Wang, Qing-Bo Liu, Xiang-Feng Yang, Hua-Hua Fu

To observe the Weyl phonon (WP) with the maximum charge and to design a realistic material structure containing only single-pair-WPs have long been considered two challenges in the field of topology physics. Here we have successfully designed an acoustic crystal to realize the single-pair-WPs with the maximum charge for the first time. Our theoretical simula

Elia Moscoso Thompson, Andrea Ranieri, Silvia Biasotti, Miguel Chicchon

This paper describes the methods submitted for evaluation to the SHREC 2022 track on pothole and crack detection in the road pavement. A total of 7 different runs for the semantic segmentation of the road surface are compared, 6 from the participants plus a baseline method. All methods exploit Deep Learning techniques and their performance is tested using th

Said Obakrim, Valérie Monbet, Nicolas Raillard, Pierre Ailliot

Ocean wave climate has a significant impact on near-shore and off-shore human activities, and its characterisation can help in the design of ocean structures such as wave energy converters and sea dikes. Therefore, engineers need long time series of ocean wave parameters. Numerical models are a valuable source of ocean wave data; however, they are computatio

Elaheh Ataeebojd, Mehdi Rasti, Hossein Pedram, Pedro H. J. Nardelli

Sharing the licensed frequency spectrum among multiple mobile network operators (MNOs) is a promising approach to improve licensed spectrum utilization. In this paper, we model and analyze a non-orthogonal spectrum sharing system consisting of multiple seller and multiple buyer MNOs where buyer MNOs lease several licensed sub-bands from different seller MNOs

Samuel J. Bell, Neil D. Lawrence

We investigate the effect of task ordering on continual learning performance. We conduct an extensive series of empirical experiments on synthetic and naturalistic datasets and show that reordering tasks significantly affects the amount of catastrophic forgetting. Connecting to the field of curriculum learning, we show that the effect of task ordering can be

S. P. Roshchupkin, S. S. Starodub

The generation of narrow beams of high-energy positrons (electrons) in the process of resonant photogeneration of ultrarelativistic electron-positron pairs by high-energy gamma quanta in the field of the nucleus and a strong electromagnetic wave is theoretically predicted. It is shown that if the energy of the initial gamma quanta significantly exceeds the c

Mayank Raikwar, Danilo Gligoroski

Denial of Service (DoS) attacks are a growing threat in network services. The frequency and intensity of DoS attacks are rapidly increasing day by day. The immense financial potential of the Cryptocurrency market is a prevalent target of the DoS attack. The DoS attack events are kept on happening in cryptocurrencies and the blockchain ecosystem. To the best

Luis A. Souto Arias, Pasquale Cirillo, Cornelis W. Oosterlee

We propose a new jump-diffusion process, the Heston-Queue-Hawkes (HQH) model, combining the well-known Heston model and the recently introduced Queue-Hawkes (Q-Hawkes) jump process. Like the Hawkes process, the HQH model can capture the effects of self-excitation and contagion. However, since the characteristic function of the HQH process is known in closed-

Yutian Chen, Xingyou Song, Chansoo Lee, Zi Wang

Meta-learning hyperparameter optimization (HPO) algorithms from prior experiments is a promising approach to improve optimization efficiency over objective functions from a similar distribution. However, existing methods are restricted to learning from experiments sharing the same set of hyperparameters. In this paper, we introduce the OptFormer, the first t

Roberto Ladu

We introduce and study a class of compact 4-manifolds with boundary that we call protocorks. Any exotic pair of simply connected closed 4-manifolds is related by a protocork twist, moreover, any cork is supported by a protocork. We prove a theorem on the relative Seiberg-Witten invariants of a protocork before and after twisting and a splitting theorem on th

Valentina Beorchia, Michela Brundu

The present paper concerns the question of the violation of the r-th inequality for extremal curves in the projective r-space, posed by T. Kato and G. Martens. We show that the answer is negative in many cases. The result is obtained by a detailed analysis of the geometry of extremal curves and their canonical model. As a consequence, we show that particular

M. Serkan Kopuzlu, M. Okan Araz, Ahmet R. Emirdagi, Murat Kuscu

Molecular Communications (MC) is a bio-inspired communication technique that uses molecules to encode and transfer information. Many efforts have been focused on developing new modulation techniques for MC by exploiting distinguishable properties of molecules. In this paper, we investigate a particular modulation scheme where the information is encoded into

Changjian Shui, Qi Chen, Jiaqi Li, Boyu Wang

We consider a fair representation learning perspective, where optimal predictors, on top of the data representation, are ensured to be invariant with respect to different sub-groups. Specifically, we formulate this intuition as a bi-level optimization, where the representation is learned in the outer-loop, and invariant optimal group predictors are updated i

Mirco Ciallella, Davide Torlo, Mario Ricchiuto

In the context of preserving stationary states, e.g. lake at rest and moving equilibria, a new formulation of the shallow water system, called Flux Globalization has been introduced by Cheng et al. (2019). This approach consists in including the integral of the source term in the global flux and reconstructing the new global flux rather than the conservative

Benjamin Metha, Michele Trenti, Tingjin Chu, Andrew Battisti

The metallicity of diffuse ionised gas (DIG) cannot be determined using strong emission line diagnostics, which are calibrated to calculate the metallicity of Hii regions. Because of this, resolved metallicity maps from integral field spectroscopy (IFS) data remain largely incomplete. In this paper (the second of a series), we introduce the geostatistical te

Sheng Guo, Zihua Xiong, Yujie Zhong, Limin Wang

In this paper, we present a new cross-architecture contrastive learning (CACL) framework for self-supervised video representation learning. CACL consists of a 3D CNN and a video transformer which are used in parallel to generate diverse positive pairs for contrastive learning. This allows the model to learn strong representations from such diverse yet meanin

Florian Girelli, Matteo Laudonio

We introduce the framework of Hopf algebra field theory (HAFT) which generalizes the notion of group field theory to the quantum group (Hopf algebra) case. We focus in particular on the 3d case and show how the HAFT we considered is topological. The highlight of the construction is the notion of plane-wave which leads, in the specific example of SUq (2) with

Neeraj Vashistha

The widely used gene quantisation technique, Lateral Flow Device (LFD), is now commonly used to detect the presence of SARS-CoV-2. It is enabling the control and prevention of the spread of the virus. Depending on the viral load, LFD have different sensitivity and self-test for normal user present additional challenge to interpret the result. With the evolut

Run Wen, FangXia An, Xian Zhong Zheng, Dong Dong Shi

We present an analysis of physical properties of 34 [O III] emission-line galaxies (ELGs) at z=3.254$\pm$0.029 in the Extended Chandra Deep Field South (ECDFS). These ELGs are selected from deep narrow H2S(1) and broad Ks imaging of 383 arcmin$^{2}$ obtained with CFHT/WIRCam. We construct spectral energy distributions (SEDs) from U to Ks to derive the physic

Antoine Deza, Mingfei Hao, Lionel Pournin

We propose a computational, convex hull free framework that takes advantage of the combinatorial structure of a zonotope, as for example its symmetry group, to orbitwise generate all canonical representatives of its vertices. We illustrate the proposed framework by generating all the 1 955 230 985 997 140 vertices of the $9$-dimensional White Whale. We also

C. Gao, C. Hofer, D. Jannis, A. Béché

Ptychography provides highly efficient imaging in scanning transmission electron microscopy (STEM), but questions have remained over its applicability to strongly scattering samples such as those most commonly seen in materialsscience. Although contrast reversals can appear in ptychographic phase images as the projected potentials of the sample increase, we

Xiao Fang, Yuta Koike

We use a new method via $p$-Wasserstein bounds to prove Cram\'er-type moderate deviations in (multivariate) normal approximations. In the classical setting that $W$ is a standardized sum of $n$ independent and identically distributed (i.i.d.) random variables with sub-exponential tails, our method recovers the optimal range of $0\leq x=o(n^{1/6})$ and the ne

Masato Tanabe

Our aim is to precisely present a tame topology counterpart to canonical stratification of a Lie groupoid. We consider a definable Lie groupoid in semialgebraic, subanalytic, o-minimal over $\mathbb{R}$, or more generally, Shiota's $\mathfrak{X}$-category. We show that there exists a canonical Whitney stratification of the Lie groupoid into definable strata

Şeyma Karadereli, Ferit Öztürk

A real algebraic link in the 3-sphere is defined as the zero locus in the 3-sphere of a real algebraic function from $\mathbb{R}^4$ to $\mathbb{R}^2$. A real algebraic open book decomposition on the 3-sphere is by definition the Milnor fibration of such a real algebraic function, in case it exists. We prove that every overtwisted contact structure on the 3-s

Debankur Bhattacharyya, Christopher Jarzynski

We present a simple strategy for constructing an information ratchet or memory-tape model of Maxwell's demon, from a feedback-controlled model. We illustrate our approach by converting the Annby-Andersson feedback-controlled double quantum dot model [Phys. Rev. B 101, 165404 (2020)] to a memory-tape model. We use the underlying network structure of the origi

Federica Gerace, Florent Krzakala, Bruno Loureiro, Ludovic Stephan

While classical in many theoretical settings - and in particular in statistical physics-inspired works - the assumption of Gaussian i.i.d. input data is often perceived as a strong limitation in the context of statistics and machine learning. In this study, we redeem this line of work in the case of generalized linear classification, a.k.a. the perceptron mo

Zhuoxuan Xie, Baoyi Chen

We treat heavy quark as an open quantum system in the hot medium and rederive the Stochastic Schr\"odinger Equation (SSE) from the full Schr\"odinger equation for both heavy quarks and the medium. We apply the SSE to the dynamical evolutions of heavy quarks (as a system) in the static hot medium (as an environment). Heavy quarks interact with the medium via

Thomas Führer, Norbert Heuer, Antti H. Niemi

We present a discontinuous Petrov-Galerkin (DPG) method with optimal test functions for the Reissner-Mindlin plate bending model. Our method is based on a variational formulation that utilizes a Helmholtz decomposition of the shear force. It produces approximations of the primitive variables and the bending moments. For any canonical selection of boundary co

Shijing Si, Jianzong Wang, Ruiyi Zhang, Qinliang Su

Non-negative matrix factorization (NMF) based topic modeling is widely used in natural language processing (NLP) to uncover hidden topics of short text documents. Usually, training a high-quality topic model requires large amount of textual data. In many real-world scenarios, customer textual data should be private and sensitive, precluding uploading to data

Zhengyang Li, Shijing Si, Jianzong Wang, Jing Xiao

Pre-trained BERT models have achieved impressive performance in many natural language processing (NLP) tasks. However, in many real-world situations, textual data are usually decentralized over many clients and unable to be uploaded to a central server due to privacy protection and regulations. Federated learning (FL) enables multiple clients collaboratively

Anthony J Blazevich, Jared R Fletcher

Elastic strain energy is stored and released from long, distal tendons such as the Achilles during locomotion, reducing locomotor energy cost by minimising muscle shortening distance and speed, and thus activation. However, numerous additional, often unrecognised, advantages of long tendons may speculatively be of greater evolutionary advantage, including th

Simon Langer, Oliver Taubmann, Felix Denzinger, Andreas Maier

Reliably detecting diseases using relevant biological information is crucial for real-world applicability of deep learning techniques in medical imaging. We debias deep learning models during training against unknown bias - without preprocessing/filtering the input beforehand or assuming specific knowledge about its distribution or precise nature in the data

Liushuai Shi, Le Wang, Chengjiang Long, Sanping Zhou

Understanding the multiple socially-acceptable future behaviors is an essential task for many vision applications. In this paper, we propose a tree-based method, termed as Social Interpretable Tree (SIT), to address this multi-modal prediction task, where a hand-crafted tree is built depending on the prior information of observed trajectory to model multiple

Suman G Das, Joachim Krug

The extent of parallel evolution at the genotypic level is quantitatively linked to the distribution of beneficial fitness effects (DBFE) of mutations. The standard view, based on light-tailed distributions (i.e. distributions with finite moments), is that the probability of parallel evolution in duplicate populations is inversely proportional to the number

Zhenpeng Feng, Milos Dakovic, Hongbing Ji, Mingzhe Zhu

Generative Adversarial Networks (GANs) can synthesize abundant photo-realistic synthetic aperture radar (SAR) images. Some recent GANs (e.g., InfoGAN), are even able to edit specific properties of the synthesized images by introducing latent codes. It is crucial for SAR image synthesis since the targets in real SAR images are with different properties due to

Qiu-Shi Zhu, Jie Zhang, Zi-Qiang Zhang, Li-Rong Dai

Speech enhancement (SE) is usually required as a front end to improve the speech quality in noisy environments, while the enhanced speech might not be optimal for automatic speech recognition (ASR) systems due to speech distortion. On the other hand, it was shown that self-supervised pre-training enables the utilization of a large amount of unlabeled noisy d

Jinbo Chen, Fengshi Tian, Jie Yang, Mohamad Sawan

Wearable electrocardiograph (ECG) recording and processing systems have been developed to detect cardiac arrhythmia to help prevent heart attacks. Conventional wearable systems, however, suffer from high energy consumption at both circuit and system levels. To overcome the design challenges, this paper proposes an event-driven compressive ECG recording and n

Manibrata Sen

Neutrinos can be pseudo-Dirac in nature -- Majorana fermions behaving as Dirac fermions for all practical purposes. In such a scenario, active and sterile neutrinos are quasi-degenerate in mass, and hence oscillations between the two, due to their tiny mass-squared difference $(\delta m^2)$, can develop only over very long baselines. Under this hypothesis, w

Yang Xiao, Dehua Liu, Jizhou He, Wu-Ming Liu

We consider a finite-time quantum Otto heat engine that consists of two isochoric (thermal-contact) process, where the system is alternatively coupled to a hot squeezed and a cold thermal reservoir, and two unitary driven strokes, where the system is isolated from these two baths and its von Neumann entropy keeps constant. Both quantum inner friction and coh

Rituraj Sharma, Nimrod Benshalom, Maor Asher, Thomas M. Brenner

$\delta$-Bi$_2$O$_3$ is one of the fastest known solid oxide ion conductors owing to its intrinsically defective fluorite-like structure with 25\% vacant sites in the O sub-lattice. Numerous diffraction measurements and molecular dynamics simulations indicate that the Bi ions construct a cubic, fcc lattice, and the O ions are %distributing and migrating thro

Krzysztof A. Mizerski

The fully developed turbulent Boussinesq convection is known to form large-scale rolls, often termed the 'large-scale circulation' (LSC). It is an interesting question how such a large-scale flow is created, in particular in systems when the energy input occurs at small scales, when inverse cascade is required in order to transfer energy into the large-scale

Akshay Kannan Sairam, C. M. Chandrashekar

Most quantum systems that are used for generating entanglement and for practical applications are not isolated from the environment, and are hence susceptible to noise. Entanglement in more than one degree of freedom between two systems, known as hyperentanglement, is known to have certain advantages, including robustness against noise over conventional enta

Rainis Haller, Andre Ostrak, Märt Põldvere

We solve some open problems regarding diameter two properties within the class of Banach spaces of real-valued Lipschitz functions by using the de Leeuw transform. Namely, we show that: the diameter two property, the strong diameter two property, and the symmetric strong diameter two property are all different for these spaces of Lipschitz functions; the spa

Leila Schneps

Let $A_\tau$ denote the elliptic associator constructed by Enriquez, a power series in two non-commutative variables $a,b$ defined as an iterated integral of the Kronecker function $F_\tau$. We study a family of {\it Fay relations} satisfied by $A_\tau$, derived from the original Fay relation satisfied by the $F_\tau$. The Fay relations of $A_\tau$ were stud

Renwang Li, Shu Sun, Meixia Tao

Reconfigurable intelligent surface (RIS) has attracted extensive attention in recent years. However, most research focuses on the scenario of the narrowband and/or instantaneous channel state information (CSI), while wide bandwidth with the use of millimeter-wave (mmWave) (including sub-Terahertz) spectrum is a major trend in next-generation wireless communi

Oliver Knill

The Babylonian graph B has the positive integers as vertices and connects two if they define a Pythagorean triple. Triangular subgraphs correspond to Euler bricks. What are the properties of this graph? Are there tetrahedral subgraphs corresponding to Euler tesseracts? Is there only one infinite connected component? Are there two Euler bricks in the graph th

Miguel Ángel González-Santamarta, Francisco Javier Rodríguez-Lera, Camino Fernández Llamas, Francisco Martín Rico

State machines are a common mechanism for defining behaviors in robots, defining them based on identifiable stages. There are several libraries available for easing the implementation of state machines in ROS 1, as SMACH or SMACC, but there are fewer alternatives for ROS 2. YASMIN is yet another library specifically designed for ROS 2 for easing the design o

Zhiwei Bai, Tao Luo, Zhi-Qin John Xu, Yaoyu Zhang

Understanding the relation between deep and shallow neural networks is extremely important for the theoretical study of deep learning. In this work, we discover an embedding principle in depth that loss landscape of an NN "contains" all critical points of the loss landscapes for shallower NNs. The key tool for our discovery is the critical lifting operator p

Yue Song, Nicu Sebe, Wei Wang

The Fine-Grained Visual Categorization (FGVC) is challenging because the subtle inter-class variations are difficult to be captured. One notable research line uses the Global Covariance Pooling (GCP) layer to learn powerful representations with second-order statistics, which can effectively model inter-class differences. In our previous conference paper, we

Varun Ravi Kumar, Ciaran Eising, Christian Witt, Senthil Yogamani

Surround-view fisheye cameras are commonly used for near-field sensing in automated driving. Four fisheye cameras on four sides of the vehicle are sufficient to cover 360{\deg} around the vehicle capturing the entire near-field region. Some primary use cases are automated parking, traffic jam assist, and urban driving. There are limited datasets and very lit

Chengyu Qiao, Zhiyu Xiang, Xinglu Wang

Visual relocalization aims to estimate the pose of a camera from one or more images. In recent years deep learning based pose regression methods have attracted many attentions. They feature predicting the absolute poses without relying on any prior built maps or stored images, making the relocalization very efficient. However, robust relocalization under env

MinGyu Choi, Wonseok Shin, Yijingxiu Lu, Sun Kim

Recent contrastive learning methods have shown to be effective in various tasks, learning generalizable representations invariant to data augmentation thereby leading to state of the art performances. Regarding the multifaceted nature of large unlabeled data used in self-supervised learning while majority of real-word downstream tasks use single format of da

Zülfiye Kütük, Görkem Algan

Semantic segmentation is a challenging task since it requires excessively more low-level spatial information of the image compared to other computer vision problems. The accuracy of pixel-level classification can be affected by many factors, such as imaging limitations and the ambiguity of object boundaries in an image. Conventional methods exploit three-cha

Duygu Sesver, Alp Eren Gençoğlu, Çağrı Emre Yıldız, Zehra Günindi

Automatic detection of natural disasters and incidents has become more important as a tool for fast response. There have been many studies to detect incidents using still images and text. However, the number of approaches that exploit temporal information is rather limited. One of the main reasons for this is that a diverse video dataset with various inciden