Datasets:

royokong

/

arxiv-abstract-23-24

like 0

2401.10111

Tuc Nguyen Van Mr

Tuc Nguyen and Thai Le

Marrying Adapters and Mixup to Efficiently Enhance the Adversarial Robustness of Pre-Trained Language Models for Text Classification

10 pages and 2 figures

cs.CL

http://creativecommons.org/licenses/by/4.0/

Existing works show that augmenting training data of neural networks using both clean and adversarial examples can enhance their generalizability under adversarial attacks. However, this training approach often leads to performance degradation on clean inputs. Additionally, it requires frequent re-training of the entire model to account for new attack types, resulting in significant and costly computations. Such limitations make adversarial training mechanisms less practical, particularly for complex Pre-trained Language Models (PLMs) with millions or even billions of parameters. To overcome these challenges while still harnessing the theoretical benefits of adversarial training, this study combines two concepts: (1) adapters, which enable parameter-efficient fine-tuning, and (2) Mixup, which train NNs via convex combinations of pairs data pairs. Intuitively, we propose to fine-tune PLMs through convex combinations of non-data pairs of fine-tuned adapters, one trained with clean and another trained with adversarial examples. Our experiments show that the proposed method achieves the best trade-off between training efficiency and predictive performance, both with and without attacks compared to other baselines on a variety of downstream tasks.

2401.10112

Jongkuk Kim

Shu-Yu Ho, Jongkuk Kim, Pyungwon Ko

Recent $\boldsymbol{B^+ \!\to K^+\nu\bar{\nu}}$ Excess and Muon $\boldsymbol{g - 2}$ Illuminating Light Dark Sector with Higgs Portal

7 pages, 3 figures

KIAS-24003

hep-ph hep-ex

http://creativecommons.org/licenses/by/4.0/

The Belle II collaboration recently announced that they observed the $\boldsymbol{B^+ \!\to K^+\nu\bar{\nu}}$ decay process for the first time. This dineutrino mode of $\boldsymbol{B^+ \!\to K^+\nu\bar{\nu}}$ has been theoretically identified as a very clean channel. However, their result encounters a $2.8{}^{}\sigma$ deviation from the Standard Model (SM) calculation. On the other hand, last year, Fermilab released new data on muon $g-2$ away from the SM expectation with $5{}^{}\sigma$. In this letter, we study the simplest UV-complete $\text{U}(1)_{\textsf{L}_\mu - \textsf{L}_\tau}^{}$-charged complex scalar Dark Matter (DM) model. Thanks to the existence of light dark Higgs boson and light dark photon, we can explain the observed relic density of DM and resolve the results reported by both Belle II and Fermilab experiments simultaneously. As a byproduct, the Hubble tension is alleviated by taking $\Delta N_\textsf{eff}^{} \simeq 0.3$ induced by the light dark photon.

2401.10113

Shan Jia

Soumyya Kanti Datta, Shan Jia, Siwei Lyu

Exposing Lip-syncing Deepfakes from Mouth Inconsistencies

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

A lip-syncing deepfake is a digitally manipulated video in which a person's lip movements are created convincingly using AI models to match altered or entirely new audio. Lip-syncing deepfakes are a dangerous type of deepfakes as the artifacts are limited to the lip region and more difficult to discern. In this paper, we describe a novel approach, LIP-syncing detection based on mouth INConsistency (LIPINC), for lip-syncing deepfake detection by identifying temporal inconsistencies in the mouth region. These inconsistencies are seen in the adjacent frames and throughout the video. Our model can successfully capture these irregularities and outperforms the state-of-the-art methods on several benchmark deepfake datasets.

2401.10114

Alessandro Bellina

Alessandro Bellina, Giordano De Marzo, Vittorio Loreto

Time-Dependent Urn Models reproduce the full spectrum of novelties discovery

cond-mat.stat-mech math.PR

http://creativecommons.org/licenses/by-nc-sa/4.0/

Systems driven by innovation, a pivotal force in human society, present various intriguing statistical regularities, from the Heaps' law to logarithmic scaling or somewhat different patterns for the innovation rates. The Urn Model with Triggering (UMT) has been instrumental in modelling these innovation dynamics. Yet, a generalisation is needed to capture the richer empirical phenomenology. Here, we introduce a Time-dependent Urn Model with Triggering (TUMT), a generalisation of the UMT that crucially integrates time-dependent parameters for reinforcement and triggering to offer a broader framework for modelling innovation in non-stationary systems. Through analytical computation and numerical simulations, we show that the TUMT reconciles various behaviours observed in a broad spectrum of systems, from patenting activity to the analysis of gene mutations. We highlight how the TUMT features a "critical" region where both Heaps' and Zipf's laws coexist, for which we compute the exponents.

2401.10115

Jonas Rabia

Jonas Rabia, Quentin N\'enon, Nicolas Andr\'e, Vincent Hue, Daniel Santos-Costa, Aneesah Kamran and Michel Blanc

Influence of the Jovian current sheet models on the mapping of the UV auroral footprints of Io, Europa, and Ganymede

22 pages, 9 figures, 1 table. Accepted for publication in Journal of Geophysical Research: Space physics

10.1029/2023JA032041

astro-ph.EP physics.space-ph

http://creativecommons.org/licenses/by/4.0/

The in-situ characterization of moon-magnetosphere interactions at Jupiter and the mapping of moon auroral footpaths require accurate global models of the magnetospheric magnetic field. In this study, we compare the ability of two widely-used current sheet models, Khurana-2005 (KK2005) and Connerney-2020 (CON2020) combined with the most recent measurements acquired at low, medium, and high latitudes. With the adjustments of the KK2005 model to JRM33, we show that in the outer and middle magnetosphere (R>15RJ), JRM33+KK2005 is found to be the best model to reproduce the magnetic field observations of Galileo and Juno as it accounts for local time effects. JRM33+CON2020 gives the most accurate representation of the inner magnetosphere. This finding is drawn from comparisons with Juno in-situ magnetic field measurements and confirmed by contrasting the timing of the crossings of the Io, Europa, and Ganymede flux tubes identified in the Juno particles data with the two model estimates. JRM33+CON2020 also maps more accurately the UV auroral footpath of Io, Europa, and Ganymede observed by Juno than JRM33+KK2005. The JRM33+KK2005 model predicts a local time asymmetry in position of the moons' footprints, which is however not detected in Juno's UV measurements.This could indicate that local time effects on the magnetic field are marginal at the orbital locations of Io, Europa, and Ganymede. Finally, the accuracy of the models and their predictions as a function of hemisphere, local time, and longitude is explored.

2401.10116

Roberto Dale

Roberto Dale and Diego S\'aez

Interpreting deviations between AR-VTG and GR

17 pages, 10 figures

Int. J. Mod. Phys. D 28 (2019) 1930002

10.1142/S0218271819300027

astro-ph.CO gr-qc

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The Cosmic microwave background (CMB) anisotropies predicted by two cosmological models are compared, one of them is the standard model of general relativity with cold dark matter and cosmological constant, whereas the second model is based on a consistent vector-tensor theory of gravitation explaining solar system and cosmological observations. It is proved that the resulting differences -- between the anisotropies of both models -- are due to the so-called late integrated Sachs Wolfe effect and, consequently, cross correlations between maps of CMB temperatures and tracers of the dark matter distribution could be used in future to select one of the above models. The role of reionization is analysed in detail.

2401.10117

Damas Mgani

Sophie Marques and Damas Mgani

A Categorical Perspective on Gluing

26 pages, 22 figures

math.CT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper introduces the concept of gluing in a general category, enabling us to define categories that admit glued-up objects. To achieve this, we introduce the notion of a gluing index category. Subsequently, we provide an entirely abstract definition of a gluing data functor requiring only the given category to admit pushouts. We explore various characterizations of cones and limits over these functors. We introduce the concept of refined gluing, which in turn enables us to combine different gluing data effectively. Furthermore, we demonstrate that several categories of topological spaces admit glued-up objects. This, in turn, allows us to establish a concept of gluing covering and to prove that the collection of those coverings forms a Grothendieck topology.

2401.10118

Alessandro Scala

Alessandro Scala

Techniques for Authenticating Quantile Digests

cs.DS cs.DC

http://creativecommons.org/licenses/by/4.0/

We investigate two possible techniques to authenticate the q-digest data structure, along with a worst-case study of the computational complexity both in time and space of the proposed solutions, and considerations on the feasibility of the presented approaches in real-world scenarios. We conclude the discussion by presenting some considerations on the information complexity of the queries in the two proposed approaches, and by presenting some interesting ideas that could be the subject of future studies on the topic.

2401.10119

Luis M\"uller

Luis M\"uller and Christopher Morris

Towards Principled Graph Transformers

cs.LG cs.AI

http://creativecommons.org/licenses/by/4.0/

Graph learning architectures based on the k-dimensional Weisfeiler-Leman (k-WL) hierarchy offer a theoretically well-understood expressive power. However, such architectures often fail to deliver solid predictive performance on real-world tasks, limiting their practical impact. In contrast, global attention-based models such as graph transformers demonstrate strong performance in practice, but comparing their expressive power with the k-WL hierarchy remains challenging, particularly since these architectures rely on positional or structural encodings for their expressivity and predictive performance. To address this, we show that the recently proposed Edge Transformer, a global attention model operating on node pairs instead of nodes, has at least 3-WL expressive power. Empirically, we demonstrate that the Edge Transformer surpasses other theoretically aligned architectures regarding predictive performance while not relying on positional or structural encodings.

2401.10120

Jeffrey Larson

Xinyu Fei and Lucas T. Brady and Jeffrey Larson and Sven Leyffer and Siqian Shen

Binary Quantum Control Optimization with Uncertain Hamiltonians

quant-ph math.OC

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Optimizing the controls of quantum systems plays a crucial role in advancing quantum technologies. The time-varying noises in quantum systems and the widespread use of inhomogeneous quantum ensembles raise the need for high-quality quantum controls under uncertainties. In this paper, we consider a stochastic discrete optimization formulation of a binary optimal quantum control problem involving Hamiltonians with predictable uncertainties. We propose a sample-based reformulation that optimizes both risk-neutral and risk-averse measurements of control policies, and solve these with two gradient-based algorithms using sum-up-rounding approaches. Furthermore, we discuss the differentiability of the objective function and prove upper bounds of the gaps between the optimal solutions to binary control problems and their continuous relaxations. We conduct numerical studies on various sized problem instances based of two applications of quantum pulse optimization; we evaluate different strategies to mitigate the impact of uncertainties in quantum systems. We demonstrate that the controls of our stochastic optimization model achieve significantly higher quality and robustness compared to the controls of a deterministic model.

2401.10121

Jeffrey Larson

Jeffrey Larson and Matt Menickelly and Jiahao Shi

A Novel Noise-Aware Classical Optimizer for Variational Quantum Algorithms

quant-ph math.OC

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

A key component of variational quantum algorithms (VQAs) is the choice of classical optimizer employed to update the parameterization of an ansatz. It is well recognized that quantum algorithms will, for the foreseeable future, necessarily be run on noisy devices with limited fidelities. Thus, the evaluation of an objective function (e.g., the guiding function in the quantum approximate optimization algorithm (QAOA) or the expectation of the electronic Hamiltonian in variational quantum eigensolver (VQE)) required by a classical optimizer is subject not only to stochastic error from estimating an expected value but also to error resulting from intermittent hardware noise. Model-based derivative-free optimization methods have emerged as popular choices of a classical optimizer in the noisy VQA setting, based on empirical studies. However, these optimization methods were not explicitly designed with the consideration of noise. In this work we adapt recent developments from the ``noise-aware numerical optimization'' literature to these commonly used derivative-free model-based methods. We introduce the key defining characteristics of these novel noise-aware derivative-free model-based methods that separate them from standard model-based methods. We study an implementation of such noise-aware derivative-free model-based methods and compare its performance on demonstrative VQA simulations to classical solvers packaged in \texttt{scikit-quant}.

2401.10122

Zhechen Li

Zhechen Li, Zimai Guo, Lirong Xia, Yongzhi Cao, Hanpin Wang

Differentially Private Approval-Based Committee Voting

cs.GT

http://creativecommons.org/licenses/by/4.0/

In this paper, we investigate tradeoffs between differential privacy (DP) and several voting axioms for approval-based committee voting, including proportionality, Pareto efficiency, Condorcet criterion, and strategyproofness. For all the axioms except strategyproofness, we show their incompatibility with DP, and provide both upper and lower bounds for their tradeoffs with DP. Furthermore, we show that any $\epsilon$-DP mechanism satisfies $e^{-\epsilon}$-cardinality strategyproofness, and the satisfaction can be further improved if the mechanism satisfies monotonicity.

2401.10123

Peter Svedlindh

Venkatesh Mottamchetty, Rimantas Brucas, Anna L. Ravensburg, Rahul Gupta, Arne Roos, Cheuk Wai Tai, Vassilios Kapaklis, and Peter Svedlindh

Terahertz emission from $\alpha$-W/CoFe epitaxial spintronic emitters

physics.app-ph

http://creativecommons.org/licenses/by/4.0/

We report efficient terahertz (THz) generation in epitaxial $\alpha$-W/Co$_6$0Fe$_4$0 spintronic emitters. Two types of emitters have been investigated; epitaxial $\alpha$-W(110)/Co$_6$0Fe$_4$0(110) and $\alpha$-W(001)/Co$_6$0Fe$_4$0(001) deposited on single crystalline Al$_2$O$_3$(11-20) and MgO(001) substrates, respectively. The generated THz radiation is about 10% larger for $\alpha$-W(110)/Co$_6$0Fe$_4$0(110) grown on single crystalline Al$_2$O$_3$(11-20), which is explained by the fact that the $\alpha$-W(110)/Co$_6$0Fe$_4$0(110) interface for this emitter is more transparent to the spin current due to the presence of Angstrom-scale interface intermixing at the W/CoFe interface. Our results also reveal that the generation of THz radiation is larger when pumping with the laser light from the substrate side, which is explained by a larger part of the laser light due to interference effects in the film stack being absorbed in the ferromagnetic Co$_6$0Fe$_4$0 layer in this measurement configuration.

2401.10124

Didong Li

Yun Jin Park, Didong Li

Lower Ricci Curvature for Efficient Community Detection

stat.ME cs.SI physics.soc-ph stat.AP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This study introduces the Lower Ricci Curvature (LRC), a novel, scalable, and scale-free discrete curvature designed to enhance community detection in networks. Addressing the computational challenges posed by existing curvature-based methods, LRC offers a streamlined approach with linear computational complexity, making it well-suited for large-scale network analysis. We further develop an LRC-based preprocessing method that effectively augments popular community detection algorithms. Through comprehensive simulations and applications on real-world datasets, including the NCAA football league network, the DBLP collaboration network, the Amazon product co-purchasing network, and the YouTube social network, we demonstrate the efficacy of our method in significantly improving the performance of various community detection algorithms.

2401.10125

V\'ictor L\'opez-Pardo

N\'estor Armesto, Elena G. Ferreiro, Miguel \'Angel Escobedo and V\'ictor L\'opez-Pardo

A potential approach to the $X(3872)$ thermal behaviour

11 pages, 7 figures

hep-ph nucl-th

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We study the potential of $X(3872)$ at finite temperature in the Born-Oppenheimer approximation under the assumption that it is a tetraquark. We argue that, at large number of colors, it is a good approximation to assume that the potential consists in a real part plus a constant imaginary term. The real part is then computed adapting an approach by Rothkopf and Lafferty and using as input lattice QCD determinations of the potential for hybrids. This model allows us to qualitatively estimate at which temperature range the formation of a heavy tetraquark is possible, and to propose a qualitative picture for the dissociation of the state in a medium. Our approach can be applied to other suggested internal structures for the $X(3872)$ and to other exotic states.

2401.10126

Simon Betzold

Simon Betzold (1), Johannes D\"ureth (1), Marco Dusel (1), Monika Emmerling (1), Antonina Bieganowska (2), J\"urgen Ohmer (3), Utz Fischer (3), Sven H\"ofling (1), and Sebastian Klembt (1) ((1) Julius-Maximilians-Universit\"at W\"urzburg, Physikalisches Institut and W\"urzburg-Dresden Cluster of Excellence ct.qmat, Lehrstuhl f\"ur Technische Physik, W\"urzburg, Germany, (2) Wroclaw University of Science and Technology, Faculty of Fundamental Problems of Technology, Department of Experimental Physics, Wroclaw, Poland, (3) Julius-Maximilians-Universit\"at W\"urzburg, Department of Biochemistry, W\"urzburg, Germany)

Observation of Dirac cones and room temperature polariton lasing in an organic honeycomb lattice

14 pages, 3 figures

physics.optics cond-mat.mes-hall

http://creativecommons.org/licenses/by/4.0/

Artificial one- and two-dimensional lattices have emerged as a powerful platform for the emulation of lattice Hamiltonians, the fundamental study of collective many-body effects as well as phenomena arising from non-trivial topology. Exciton-polaritons, bosonic part-light and part-matter quasiparticles, combine pronounced nonlinearities with the possibility of on-chip implementation. In this context, organic semiconductors, hosting ultra-stable Frenkel excitons, embedded in a microcavity have proven to be versatile contenders for the study of nonlinear many-body physics and bosonic condensation, which also enable deployment at ambient conditions. Here, we implement a well-controlled, high-quality optical lattice, hosting light-matter quasi-particles. The realized polariton graphene presents with excellent cavity quality factors, showing distinct signatures of Dirac cone and flatband dispersions as well as polariton lasing at room temperature. This is realized by filling coupled dielectric microcavities with the fluorescent protein mCherry. We demonstrate the emergence of a coherent polariton condensate at ambient conditions, profiting from coupling conditions as precise and controllable as in state-of-the-art inorganic semiconductor-based systems, without limitations due to e.g. lattice matching in epitaxial growth. This progress allows straightforward extension to more complex systems, such as the study of topological phenomena in two-dimensional lattices including topological lasers and non-Hermitian optics.

2401.10127

Emanuele Roberto Nocera

Felix Hekhorn, Giacomo Magni, Emanuele R. Nocera, Tanjona R. Rabemananjara, Juan Rojo, Adrianne Schaus, Roy Stegeman

Heavy Quarks in Polarised Deep-Inelastic Scattering at the Electron-Ion Collider

24 pages, 15 figures

Nikhef-2023-010, TIF-UNIMI-2023-27, Edinburgh 2023/2

hep-ph hep-ex nucl-ex nucl-th

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We extend the FONLL general-mass variable-flavour-number scheme to the case of longitudinally polarised DIS structure functions, accounting for perturbative corrections up to $\mathcal{O}(\alpha_s^2)$. We quantify the impact of charm quark mass and higher-order perturbative corrections on projected measurements of inclusive and charm-tagged longitudinal asymmetries at the Electron-Ion Collider (EIC) and at the Electron-ion collider in China (EicC). We demonstrate how the inclusion of these corrections is essential to compute predictions with an accuracy that matches the projected precision of the measurements. The computation is made publicly available through the open-source EKO and YADISM programs

2401.10128

Shaohua Pi

Lingyun Wang, Jose A Sahel, Shaohua Pi

Sub2Full: split spectrum to boost OCT despeckling without clean data

eess.IV cs.CV

http://creativecommons.org/licenses/by/4.0/

Optical coherence tomography (OCT) suffers from speckle noise, causing the deterioration of image quality, especially in high-resolution modalities like visible light OCT (vis-OCT). The potential of conventional supervised deep learning denoising methods is limited by the difficulty of obtaining clean data. Here, we proposed an innovative self-supervised strategy called Sub2Full (S2F) for OCT despeckling without clean data. This approach works by acquiring two repeated B-scans, splitting the spectrum of the first repeat as a low-resolution input, and utilizing the full spectrum of the second repeat as the high-resolution target. The proposed method was validated on vis-OCT retinal images visualizing sublaminar structures in outer retina and demonstrated superior performance over conventional Noise2Noise and Noise2Void schemes. The code is available at https://github.com/PittOCT/Sub2Full-OCT-Denoising.

2401.10129

Marcelo Saval Calvo

Alejandro Gal\'an-Cuenca, Antonio Javier Gallego, Marcelo Saval-Calvo, Antonio Pertusa

Few-shot learning for COVID-19 Chest X-Ray Classification with Imbalanced Data: An Inter vs. Intra Domain Study

Submited to Pattern Analysis and Applications

eess.IV cs.CV

http://creativecommons.org/licenses/by-nc-nd/4.0/

Medical image datasets are essential for training models used in computer-aided diagnosis, treatment planning, and medical research. However, some challenges are associated with these datasets, including variability in data distribution, data scarcity, and transfer learning issues when using models pre-trained from generic images. This work studies the effect of these challenges at the intra- and inter-domain level in few-shot learning scenarios with severe data imbalance. For this, we propose a methodology based on Siamese neural networks in which a series of techniques are integrated to mitigate the effects of data scarcity and distribution imbalance. Specifically, different initialization and data augmentation methods are analyzed, and four adaptations to Siamese networks of solutions to deal with imbalanced data are introduced, including data balancing and weighted loss, both separately and combined, and with a different balance of pairing ratios. Moreover, we also assess the inference process considering four classifiers, namely Histogram, $k$NN, SVM, and Random Forest. Evaluation is performed on three chest X-ray datasets with annotated cases of both positive and negative COVID-19 diagnoses. The accuracy of each technique proposed for the Siamese architecture is analyzed separately and their results are compared to those obtained using equivalent methods on a state-of-the-art CNN. We conclude that the introduced techniques offer promising improvements over the baseline in almost all cases, and that the selection of the technique may vary depending on the amount of data available and the level of imbalance.

2401.10130

Mattia Cafasso

Mattia Cafasso, Tom Claeys

Biorthogonal measures, polymer partition functions, and random matrices

46 pages, 3 figures, 1 table

math-ph math.CA math.CV math.MP math.PR

http://creativecommons.org/licenses/by/4.0/

We develop the study of a particular class of biorthogonal measures, encompassing at the same time several random matrix models and partition functions of polymers. This general framework allows us to characterize the partition functions of the Log Gamma polymer and the mixed polymer in terms of explicit biorthogonal measures, as it was previously done for the homogeneous O'Connell-Yor polymer by Imamura and Sasamoto. In addition, we show that the biorthogonal measures associated to these three polymer models (Log Gamma, O'Connell-Yor, and mixed polymer) converge to random matrix eigenvalue distributions in small temperature limits. We also clarify the connection between different Fredholm determinant representations and explain how our results might be useful for asymptotic analysis and large deviation estimates.

2401.10131

Ava Nederlander

Ava Nederlander, Adele Plunkett, Antonio Hales, \'Agnes K\'osp\'al, Jacob A. White, Makoto A. Johnstone, M\'aria Kun, P\'eter \'Abrah\'am, and Anna G. Hughes

An Outbursting Protostar: The environment of L1251 VLA 6

16 pages, 9 figures, 5 tables, Accepted in ApJ

astro-ph.SR

http://creativecommons.org/licenses/by/4.0/

Young protostars that undergo episodic accretion can provide insight into the impact on their circumstellar environments while matter is accreted from the disk onto the protostar. IRAS 22343+7501 is a four component protostar system with one of those being a fading outbursting protostar referred to as L1251 VLA 6. Given the rarity of YSOs undergoing this type of accretion, L1251 VLA 6 can elucidate the fading phase of the post-outburst process. Here we examine structure in the disk around L1251 VLA 6 at frequencies of 10 GHz and 33 GHz with the Karl G. Jansky Very Large Array (VLA). We model the disk structure using Markov chain Monte Carlo (MCMC). This method is then combined with a parametric ray-tracing code to generate synthetic model images of an axisymmetric disk, allowing us to characterize the radial distribution of dust in the system. The results of our MCMC fit show that the most probable values for the mass and radius are consistent with values typical of Class I objects. We find that the total mass of the disk is $0.070^{+0.031}_{-0.2} \rm ~ M_{\sun}$ and investigate the conditions that could cause the accretion outburst. We conclude that the eruption is not caused by gravitational instability and consider alternative explanations and trigger mechanisms.

2401.10132

Andrew Voitiv

Andrew A. Voitiv and Mark E. Siemens and Mark T. Lusk

Trapped Vortex Dynamics Implemented in Composite Bessel Beams

Physics of Fluids 36, 016135 (2024)

10.1063/5.0187870

physics.optics cond-mat.quant-gas physics.flu-dyn

http://creativecommons.org/licenses/by/4.0/

The divergence-free nature of Bessel beams can be harnessed to effectively trap optical vortices in free space laser propagation. We show how to generate arbitrary vortex configurations in Bessel traps to investigate few-body vortex interactions within a dynamically-evolving fluid of light, which is a formal analog to a non-interacting Bose gas. We implement--theoretically and experimentally--initial conditions of vortex configurations first predicted in harmonically-trapped quantum fluids, in the limit of weak atomic interactions, and model and measure the resultant dynamics. These hard trap dynamics are distinct from the harmonic trap predictions due to the non-local interactions that occur among the hard wall boundary and steep phase gradients that nucleate other vortices. By simultaneously presenting experimental demonstrations with the theoretical proposal, we validate the potential application of using Bessel hard wall traps as testing grounds for engineering few-body vortex interactions within trapped, two-dimensional compressible fluids.

2401.10133

Zinat Behdad

Zinat Behdad, \"Ozlem Tu\u{g}fe Demir, Ki Won Sung, Cicek Cavdar

Interplay between Sensing and Communication in Cell-Free Massive MIMO with URLLC Users

6 pages, 3 figures

cs.IT eess.SP math.IT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper studies integrated sensing and communication (ISAC) in the downlink of a cell-free massive multiple-input multiple-output (MIMO) system with multi-static sensing and ultra-reliable low-latency communication (URLLC) users. We propose a successive convex approximation-based power allocation algorithm that maximizes energy efficiency while satisfying the sensing and URLLC requirements. In addition, we provide a new definition for network availability, which accounts for both sensing and URLLC requirements. The impact of blocklength, sensing requirement, and required reliability as a function of decoding error probability on network availability and energy efficiency is investigated. The proposed power allocation algorithm is compared to a communication-centric approach where only the URLLC requirement is considered. It is shown that the URLLC-only approach is incapable of meeting sensing requirements, while the proposed ISAC algorithm fulfills both sensing and URLLC requirements, albeit with an associated increase in energy consumption. This increment can be reduced up to 75% by utilizing additional symbols for sensing. It is also demonstrated that larger blocklengths enhance network availability and offer greater robustness against stringent reliability requirements.

2401.10134

Chenxi Liu

Chenxi Liu, Sun Yang, Qianxiong Xu, Zhishuai Li, Cheng Long, Ziyue Li, Rui Zhao

Spatial-Temporal Large Language Model for Traffic Prediction

cs.LG cs.CL

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Traffic prediction, a critical component for intelligent transportation systems, endeavors to foresee future traffic at specific locations using historical data. Although existing traffic prediction models often emphasize developing complex neural network structures, their accuracy has not seen improvements accordingly. Recently, Large Language Models (LLMs) have shown outstanding capabilities in time series analysis. Differing from existing models, LLMs progress mainly through parameter expansion and extensive pre-training while maintaining their fundamental structures. In this paper, we propose a Spatial-Temporal Large Language Model (ST-LLM) for traffic prediction. Specifically, ST-LLM redefines the timesteps at each location as tokens and incorporates a spatial-temporal embedding module to learn the spatial location and global temporal representations of tokens. Then these representations are fused to provide each token with unified spatial and temporal information. Furthermore, we propose a novel partially frozen attention strategy of the LLM, which is designed to capture spatial-temporal dependencies for traffic prediction. Comprehensive experiments on real traffic datasets offer evidence that ST-LLM outperforms state-of-the-art models. Notably, the ST-LLM also exhibits robust performance in both few-shot and zero-shot prediction scenarios.

2401.10135

Raphael Schoof

Raphael Schoof, Lennart Fl\"ur, Florian Tuschner, Willy D\"orfler

Residual Based Error Estimator for Chemical-Mechanically Coupled Battery Active Particles

math.NA cs.NA

http://creativecommons.org/licenses/by-nc-sa/4.0/

Adaptive finite element methods are a powerful tool to obtain numerical simulation results in a reasonable time. Due to complex chemical and mechanical couplings in lithium-ion batteries, numerical simulations are very helpful to investigate promising new battery active materials such as amorphous silicon featuring a higher energy density than graphite. Based on a thermodynamically consistent continuum model with large deformation and chemo-mechanically coupled approach, we compare three different spatial adaptive refinement strategies: Kelly-, gradient recovery- and residual based error estimation. For the residual based case, the strong formulation of the residual is explicitly derived. With amorphous silicon as example material, we investigate two 3D representative host particle geometries, reduced with symmetry assumptions to a 1D unit interval and a 2D elliptical domain. Our numerical studies show that the Kelly estimator overestimates the error, whereas the gradient recovery estimator leads to lower refinement levels and a good capture of the change of the lithium flux. The residual based error estimator reveals a strong dependency on the cell error part which can be improved by a more suitable choice of constants to be more efficient. In a 2D domain, the concentration has a larger influence on the mesh distribution than the Cauchy stress.

2401.10136

Audris Mockus

Addi Malviya-Thakur, Audris Mockus

The Role of Data Filtering in Open Source Software Ranking and Selection

International Workshop on Methodological Issues with Empirical Studies in Software Engineering (WSESE 2024)

cs.SE

http://creativecommons.org/licenses/by-nc-nd/4.0/

Faced with over 100M open source projects most empirical investigations select a subset. Most research papers in leading venues investigated filtering projects by some measure of popularity with explicit or implicit arguments that unpopular projects are not of interest, may not even represent "real" software projects, or that less popular projects are not worthy of study. However, such filtering may have enormous effects on the results of the studies if and precisely because the sought-out response or prediction is in any way related to the filtering criteria. We exemplify the impact of this practice on research outcomes: how filtering of projects listed on GitHub affects the assessment of their popularity. We randomly sample over 100,000 repositories and use multiple regression to model the number of stars (a proxy for popularity) based on the number of commits, the duration of the project, the number of authors, and the number of core developers. Comparing control with the entire dataset with a filtered model projects having ten or more authors we find that while certain characteristics of the repository consistently predict popularity, the filtering process significantly alters the relation ships between these characteristics and the response. The number of commits exhibited a positive correlation with popularity in the control sample but showed a negative correlation in the filtered sample. These findings highlight the potential biases introduced by data filtering and emphasize the need for careful sample selection in empirical research of mining software repositories. We recommend that empirical work should either analyze complete datasets such as World of Code, or employ stratified random sampling from a complete dataset to ensure that filtering is not biasing the results.

2401.10137

Satoshi Yoshida

Satoshi Yoshida, Akihito Soeda, Mio Murao

Universal adjointation of isometry operations using transformation of quantum supermaps

65 pages, 6 figures

quant-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The full characterization of the possible transformations of quantum operations is indispensable to developing algorithms in higher-order quantum computation, which is the quantum version of functional programming. Although universal transformations of unitary operations have been well investigated, their extensions to non-unitary operations are still missing, except for a few examples. Here we construct \emph{isometry adjointation} protocols, transforming an input isometry operation into its adjoint operation. This task reduces to the transformation of unitary operation or that of quantum states in special cases. Parallel and sequential isometry adjointation protocols are constructed by transforming unitary inversion protocols using the composition of quantum combs. This construction achieves the optimal approximation error, which implies that the optimal performance does not depend on the output dimension of the isometry operation. In particular, we explicitly obtain the asymptotically optimal parallel protocol achieving the approximation error $\epsilon = \Theta(d^2/n)$, where $d$ is the input dimension of the isometry operation and $n$ is the number of calls of the isometry operation. We also construct the protocols for the related tasks called isometry inversion and universal error detection. We conduct semidefinite programming to investigate the optimal performances of the tasks using general protocols including indefinite causal order protocols. The numerical results show that the optimal performances of general protocols do not depend on the output dimension of the isometry operation for isometry adjointation and universal error detection, which is shown analytically for parallel and sequential protocols. They also exhibit the advantage of indefinite causal order protocols over sequential protocols for isometry inversion and universal error detection.

2401.10138

Hyunuk Kim

Seonbin Jo, Woo-Sung Jung, Hyunuk Kim

Wallets' explorations across non-fungible token collections

physics.soc-ph

http://creativecommons.org/licenses/by-nc-nd/4.0/

Non-fungible tokens (NFTs), which are immutable and transferable tokens on blockchain networks, have been used to certify the ownership of digital images often grouped in collections. Depending on individual interests, wallets explore and purchase NFTs in one or more image collections. Among many potential factors of shaping purchase trajectories, this paper specifically examines how visual similarities between collections affect wallets' explorations. Our model shows that wallets' explorations are not random but tend to favor collections having similar visual features to their previous purchases. The model also predicts the extent to which the next collection is close to the most recent collection of purchases with respect to visual features. These results are expected to enhance and support recommendation systems for the NFT market.

2401.10139

Pedro David Carneiro Neto

Eduarda Caldeira, Pedro C. Neto, Marco Huber, Naser Damer, Ana F. Sequeira

Model Compression Techniques in Biometrics Applications: A Survey

Under review at IEEE Journal

cs.CV cs.AI

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The development of deep learning algorithms has extensively empowered humanity's task automatization capacity. However, the huge improvement in the performance of these models is highly correlated with their increasing level of complexity, limiting their usefulness in human-oriented applications, which are usually deployed in resource-constrained devices. This led to the development of compression techniques that drastically reduce the computational and memory costs of deep learning models without significant performance degradation. This paper aims to systematize the current literature on this topic by presenting a comprehensive survey of model compression techniques in biometrics applications, namely quantization, knowledge distillation and pruning. We conduct a critical analysis of the comparative value of these techniques, focusing on their advantages and disadvantages and presenting suggestions for future work directions that can potentially improve the current methods. Additionally, we discuss and analyze the link between model bias and model compression, highlighting the need to direct compression research toward model fairness in future works.

2401.10140

Burkhard Schmidt

Leonardo Araujo and Caroline Lasser and Burkhard Schmidt

FSSH-2: Fewest Switches Surface Hopping with robust switching probability

physics.comp-ph

http://creativecommons.org/licenses/by/4.0/

This study introduces the FSSH-2 scheme, a redefined and numerically stable adiabatic Fewest Switches Surface Hopping (FSSH) method for mixed quantum-classical dynamics. It reformulates the standard FSSH hopping probability without non-adiabatic coupling vectors and allows for numerical time integration with larger step sizes. The advantages of FSSH-2 are demonstrated by numerical experiments for five different model systems in one and two spatial dimensions with up to three electronic states.

2401.10141

Jannis K\"orner

Anton Arnold, Christian Klein, Jannis K\"orner, Jens Markus Melenk

Optimally truncated WKB approximation for the 1D stationary Schr\"odinger equation in the highly oscillatory regime

38 pages, 10 figures

math.NA cs.NA

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

This paper is dedicated to the efficient numerical computation of solutions to the 1D stationary Schr\"odinger equation in the highly oscillatory regime. We compute an approximate solution based on the well-known WKB-ansatz, which relies on an asymptotic expansion w.r.t. the small parameter $\varepsilon$. Assuming that the coefficient in the equation is analytic, we derive an explicit error estimate for the truncated WKB series, in terms of $\varepsilon$ and the truncation order $N$. For any fixed $\varepsilon$, this allows to determine the optimal truncation order $N_{opt}$ which turns out to be proportional to $\varepsilon^{-1}$. When chosen this way, the resulting error of the optimally truncated WKB series behaves like $\mathcal{O}(\varepsilon^{-2}\exp(-r/\varepsilon))$, with some parameter $r>0$. The theoretical results established in this paper are confirmed by several numerical examples.

2401.10142

Roy Garcia

Roy J. Garcia, Kaifeng Bu, Liyuan Chen, Anton M. Graf

Resource Theory of Non-Revivals with Applications to Quantum Many-Body Scars

7+19 pages

quant-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The study of state revivals has a long history in dynamical systems. We introduce a resource theory to understand the use of state revivals in quantum physics, especially in quantum many-body scarred systems. In this theory, a state is said to contain no amount of resource if it experiences perfect revivals under some unitary evolution. All other states are said to be resourceful. We show that this resource bounds information scrambling. Furthermore, we show that quantum many-body scarred dynamics can produce revivals in the Hayden-Preskill decoding protocol and can also be used to recover damaged quantum information. Our theory establishes a framework to study information retrieval and its applications in quantum many-body physics.

2401.10143

Krishna Balajirao Manoorkar

Ineke van der Berg, Andrea De Domenico, Giuseppe Greco, Krishna Manoorkar, Alessandra Palmigiano, Mattia Panettiere

Labelled calculi for the logics of rough concepts

math.LO

http://creativecommons.org/licenses/by/4.0/

We introduce sound and complete labelled sequent calculi for the basic normal non-distributive modal logic L and some of its axiomatic extensions, where the labels are atomic formulas of the first order language of enriched formal contexts, i.e., relational structures based on formal contexts which provide complete semantics for these logics. We also extend these calculi to provide a proof system for the logic of rough formal contexts.

2401.10144

Yiqun Lin

Yiqun Lin, Liang Pan, Yi Li, Ziwei Liu, and Xiaomeng Li

Exploiting Hierarchical Interactions for Protein Surface Learning

Accepted to J-BHI

q-bio.BM cs.LG

http://creativecommons.org/licenses/by/4.0/

Predicting interactions between proteins is one of the most important yet challenging problems in structural bioinformatics. Intrinsically, potential function sites in protein surfaces are determined by both geometric and chemical features. However, existing works only consider handcrafted or individually learned chemical features from the atom type and extract geometric features independently. Here, we identify two key properties of effective protein surface learning: 1) relationship among atoms: atoms are linked with each other by covalent bonds to form biomolecules instead of appearing alone, leading to the significance of modeling the relationship among atoms in chemical feature learning. 2) hierarchical feature interaction: the neighboring residue effect validates the significance of hierarchical feature interaction among atoms and between surface points and atoms (or residues). In this paper, we present a principled framework based on deep learning techniques, namely Hierarchical Chemical and Geometric Feature Interaction Network (HCGNet), for protein surface analysis by bridging chemical and geometric features with hierarchical interactions. Extensive experiments demonstrate that our method outperforms the prior state-of-the-art method by 2.3% in site prediction task and 3.2% in interaction matching task, respectively. Our code is available at https://github.com/xmed-lab/HCGNet.

2401.10145

Jos\'e Manuel Peula-Garc\'ia

Teresa Del Castillo-Santaella, Jose Manuel Peula Garcia, Julia Maldonado-Valderrama, Ana Belen Jodar-Reyes

Interaction of surfactant and protein at the o/w interface and its effect on colloidal and biological properties of polymeric nanocarriers

8 pages, 4 figures, 3 tables

Colloids and Surfaces B: Biointerfaces, 173 (2019) 295

10.1016/j.colsurfb.2018.09.072

cond-mat.soft physics.app-ph

http://creativecommons.org/licenses/by-nc-nd/4.0/

The use of polymer-based surfactants in the double-emulsion (water/oil/water, W/O/W) solvent-evaporation technique is becoming a widespread strategy for preparing biocompatible and biodegradable polymeric nanoparticles (NPs) loaded with biomolecules of interest in biomedicine, or biotechnology. This approach enhances the stability of the NPs, reduces their size and recognition by the mononuclear phagocytic system, and protects the encapsulated biomolecule against losing biological activity. Different protocols to add the surfactant during the synthesis lead to different NP colloidal properties and biological activity.

2401.10146

Pat Lank

Pat Lank, Noah Olander

Approximation by perfect complexes detects Rouquier dimension

v1, comments welcome!

math.AG math.AC

http://creativecommons.org/licenses/by/4.0/

This work explores bounds on the Rouquier dimension in the bounded derived category of coherent sheaves on Noetherian schemes. By utilizing approximations, we exhibit that Rouquier dimension is inherently characterized by the number of cones required to build all perfect complexes. We use this to prove sharper bounds on Rouquier dimension of singular schemes. Firstly, we show Rouquier dimension doesn't go up along \'{e}tale extensions and is invariant under \'{e}tale covers of affine schemes admitting a dualizing complex. Secondly, we demonstrate that the Rouquier dimension of the bounded derived category for a curve, with a delta invariant of at most one at closed points, is no larger than two. Thirdly, we bound the Rouquier dimension for the bounded derived category of a (birational) derived splinter variety by that of a resolution of singularities.

2401.10147

Andreas Bluhm

Andreas Bluhm, \'Angela Capel, Antonio P\'erez-Hern\'andez

Strong decay of correlations for Gibbs states in any dimension

37 pages, 4 figures

quant-ph math-ph math.MP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Quantum systems in thermal equilibrium are described using Gibbs states. The correlations in such states determine how difficult it is to describe or simulate them. In this article, we show that systems with short-range interactions that are above a critical temperature satisfy a mixing condition, that is that for any regions $A$, $C$ the distance of the reduced state $\rho_{AC}$ on these regions to the product of its marginals, $$\| \rho_{AC} \rho_A^{-1} \otimes \rho_C^{-1} - \mathbf{1}_{AC}\| \, ,$$ decays exponentially with the distance between regions $A$ and $C$. This mixing condition is stronger than other commonly studied measures of correlation. In particular, it implies the exponential decay of the mutual information between distant regions. The mixing condition has been used, for example, to prove positive log-Sobolev constants. On the way, we investigate the relations to other notions of decay of correlations in quantum many-body systems and show that many of them are equivalent under the assumption that there exists a local effective Hamiltonian. The proof employs a variety of tools such as Araki's expansionals and quantum belief propagation.

2401.10148

Riccardo Majellaro

Riccardo Majellaro, Jonathan Collu, Aske Plaat, Thomas M. Moerland

Explicitly Disentangled Representations in Object-Centric Learning

cs.CV cs.AI cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Extracting structured representations from raw visual data is an important and long-standing challenge in machine learning. Recently, techniques for unsupervised learning of object-centric representations have raised growing interest. In this context, enhancing the robustness of the latent features can improve the efficiency and effectiveness of the training of downstream tasks. A promising step in this direction is to disentangle the factors that cause variation in the data. Previously, Invariant Slot Attention disentangled position, scale, and orientation from the remaining features. Extending this approach, we focus on separating the shape and texture components. In particular, we propose a novel architecture that biases object-centric models toward disentangling shape and texture components into two non-overlapping subsets of the latent space dimensions. These subsets are known a priori, hence before the training process. Experiments on a range of object-centric benchmarks reveal that our approach achieves the desired disentanglement while also numerically improving baseline performance in most cases. In addition, we show that our method can generate novel textures for a specific object or transfer textures between objects with distinct shapes.

2401.10149

Alec Wilson

Alec Wilson, Ryan Menzies, Neela Morarji, David Foster, Marco Casassa Mont, Esin Turkbeyler, Lisa Gralewski

Multi-Agent Reinforcement Learning for Maritime Operational Technology Cyber Security

13 pages, 7 figures, Proceedings of the Conference on Applied Machine Learning in Information Security 2023 (CAMLIS)

cs.LG cs.CR cs.MA

http://creativecommons.org/licenses/by/4.0/

This paper demonstrates the potential for autonomous cyber defence to be applied on industrial control systems and provides a baseline environment to further explore Multi-Agent Reinforcement Learning's (MARL) application to this problem domain. It introduces a simulation environment, IPMSRL, of a generic Integrated Platform Management System (IPMS) and explores the use of MARL for autonomous cyber defence decision-making on generic maritime based IPMS Operational Technology (OT). OT cyber defensive actions are less mature than they are for Enterprise IT. This is due to the relatively brittle nature of OT infrastructure originating from the use of legacy systems, design-time engineering assumptions, and lack of full-scale modern security controls. There are many obstacles to be tackled across the cyber landscape due to continually increasing cyber-attack sophistication and the limitations of traditional IT-centric cyber defence solutions. Traditional IT controls are rarely deployed on OT infrastructure, and where they are, some threats aren't fully addressed. In our experiments, a shared critic implementation of Multi Agent Proximal Policy Optimisation (MAPPO) outperformed Independent Proximal Policy Optimisation (IPPO). MAPPO reached an optimal policy (episode outcome mean of 1) after 800K timesteps, whereas IPPO was only able to reach an episode outcome mean of 0.966 after one million timesteps. Hyperparameter tuning greatly improved training performance. Across one million timesteps the tuned hyperparameters reached an optimal policy whereas the default hyperparameters only managed to win sporadically, with most simulations resulting in a draw. We tested a real-world constraint, attack detection alert success, and found that when alert success probability is reduced to 0.75 or 0.9, the MARL defenders were still able to win in over 97.5% or 99.5% of episodes, respectively.

2401.10150

Changgu Chen

Changgu Chen, Junwei Shu, Lianggangxu Chen, Gaoqi He, Changbo Wang and Yang Li

Motion-Zero: Zero-Shot Moving Object Control Framework for Diffusion-Based Video Generation

9 pages, 4 figures, IJCAI paper

cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Recent large-scale pre-trained diffusion models have demonstrated a powerful generative ability to produce high-quality videos from detailed text descriptions. However, exerting control over the motion of objects in videos generated by any video diffusion model is a challenging problem. In this paper, we propose a novel zero-shot moving object trajectory control framework, Motion-Zero, to enable a bounding-box-trajectories-controlled text-to-video diffusion model.To this end, an initial noise prior module is designed to provide a position-based prior to improve the stability of the appearance of the moving object and the accuracy of position. In addition, based on the attention map of the U-net, spatial constraints are directly applied to the denoising process of diffusion models, which further ensures the positional and spatial consistency of moving objects during the inference. Furthermore, temporal consistency is guaranteed with a proposed shift temporal attention mechanism. Our method can be flexibly applied to various state-of-the-art video diffusion models without any training process. Extensive experiments demonstrate our proposed method can control the motion trajectories of objects and generate high-quality videos.

2401.10151

Joa Weber

Urs Frauenfelder and Joa Weber

Local gluing

45 pages, 3 figures

math.SG math.CA math.DG math.DS math.GT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In the local gluing one glues local neighborhoods around the critical point of the stable and unstable manifolds to gradient flow lines defined on a finite time interval $[-T,T]$ for large $T$. If the Riemannian metric around the critical point is locally Euclidean, the local gluing map can be written down explicitly. In the non-Euclidean case the construction of the local gluing map requires an intricate version of the implicit function theorem. In this paper we explain a functional analytic approach how the local gluing map can be defined. For that we are working on infinite dimensional path spaces and also interpret stable and unstable manifolds as submanifolds of path spaces. The advantage of this approach is that similar functional analytical techniques can as well be generalized to infinite dimensional versions of Morse theory, for example Floer theory. A crucial ingredient is the Newton-Picard map. We work out an abstract version of it which does not involve troublesome quadratic estimates.

2401.10152

Stefan Steinerberger

Stefan Steinerberger

Sums of square roots that are close to an integer

math.NT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Let $k \in \mathbb{N}$ and suppose we are given $k$ integers $1 \leq a_1, \dots, a_k \leq n$. If $\sqrt{a_1} + \dots + \sqrt{a_k}$ is not an integer, how close can it be to one? When $k=1$, the distance to the nearest integer is $\gtrsim n^{-1/2}$. Angluin-Eisenstat observed the bound $\gtrsim n^{-3/2}$ when $k=2$. We prove there is a universal $c>0$ such that, for all $k \geq 2$, there exists a $c_k > 0$ and $k$ integers in $\left\{1,2,\dots, n\right\}$ with $$ 0 <\|\sqrt{a_1} + \dots + \sqrt{a_k} \| \leq c_k\cdot n^{-c \cdot k^{1/3}},$$ where $\| \cdot \|$ denotes the distance to the nearest integer. This is a case of the square-root sum problem in numerical analysis where the usual cancellation constructions do not apply: even for $k=3$, constructing explicit examples of integers whose square root sum is nearly an integer appears to be nontrivial.

2401.10153

Haonan Tong

Jie Lv, Haonan Tong, Qiang Pan, Zhilong Zhang, Xinxin He, Tao Luo, Changchuan Yin

Importance-Aware Image Segmentation-based Semantic Communication for Autonomous Driving

10 pages, 8 figures

cs.NI cs.CV

http://creativecommons.org/licenses/by-nc-sa/4.0/

This article studies the problem of image segmentation-based semantic communication in autonomous driving. In real traffic scenes, detecting the key objects (e.g., vehicles, pedestrians and obstacles) is more crucial than that of other objects to guarantee driving safety. Therefore, we propose a vehicular image segmentation-oriented semantic communication system, termed VIS-SemCom, where image segmentation features of important objects are transmitted to reduce transmission redundancy. First, to accurately extract image semantics, we develop a semantic codec based on Swin Transformer architecture, which expands the perceptual field thus improving the segmentation accuracy. Next, we propose a multi-scale semantic extraction scheme via assigning the number of Swin Transformer blocks for diverse resolution features, thus highlighting the important objects' accuracy. Furthermore, the importance-aware loss is invoked to emphasize the important objects, and an online hard sample mining (OHEM) strategy is proposed to handle small sample issues in the dataset. Experimental results demonstrate that the proposed VIS-SemCom can achieve a coding gain of nearly 6 dB with a 60% mean intersection over union (mIoU), reduce the transmitted data amount by up to 70% with a 60% mIoU, and improve the segmentation intersection over union (IoU) of important objects by 4%, compared to traditional transmission scheme.

2401.10154

Wesley Calvert

Wesley Calvert, Valentina Harizanov, Eugenio G. Omodeo, Alberto Policriti, and Alexandra Shlapentokh

In Memory of Martin Davis

math.HO cs.LO math.LO math.NT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The present paper gives an account for the general mathematical reader of the life and work of Martin Davis. Since two rather comprehensive autobiographical accounts and two long biographical interviews already exist, the present work focusses on Davis's scientific achievements, including work on computably enumerable sets, universal Turing machines, the hyperarithmetical hierarchy, neural networks, Hilbert's Tenth Problem, and automated reasoning.

2401.10155

Dai Benao

Ben Ao Dai, Bao-Lin Ye

A novel hybrid time-varying graph neural network for traffic flow forecasting

12 pages 1figures

cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Real-time and accurate traffic flow prediction is the foundation for ensuring the efficient operation of intelligent transportation systems.In existing traffic flow prediction methods based on graph neural networks (GNNs), pre-defined graphs were usually used to describe the spatial correlations of different traffic nodes in urban road networks. However, the ability of pre-defined graphs used to describe spatial correlation was limited by prior knowledge and graph generation methods. Although time-varying graphs based on data-driven learning can partially overcome the drawbacks of pre-defined graphs, the learning ability of existing adaptive graphs was limited. For example, time-varying graphs cannot adequately capture the inherent spatial correlations in traffic flow data.In order to solve these problems, we have proposed a hybrid time-varying graph neural network (HTVGNN) for traffic flow prediction.

2401.10156

Kaige Qu

Kaige Qu, Weihua Zhuang, Qiang Ye, Wen Wu, and Xuemin Shen

Model-Assisted Learning for Adaptive Cooperative Perception of Connected Autonomous Vehicles

Accepted by IEEE Transactions on Wireless Communications

cs.NI eess.SP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Cooperative perception (CP) is a key technology to facilitate consistent and accurate situational awareness for connected and autonomous vehicles (CAVs). To tackle the network resource inefficiency issue in traditional broadcast-based CP, unicast-based CP has been proposed to associate CAV pairs for cooperative perception via vehicle-to-vehicle transmission. In this paper, we investigate unicast-based CP among CAV pairs. With the consideration of dynamic perception workloads and channel conditions due to vehicle mobility and dynamic radio resource availability, we propose an adaptive cooperative perception scheme for CAV pairs in a mixed-traffic autonomous driving scenario with both CAVs and human-driven vehicles. We aim to determine when to switch between cooperative perception and stand-alone perception for each CAV pair, and allocate communication and computing resources to cooperative CAV pairs for maximizing the computing efficiency gain under perception task delay requirements. A model-assisted multi-agent reinforcement learning (MARL) solution is developed, which integrates MARL for an adaptive CAV cooperation decision and an optimization model for communication and computing resource allocation. Simulation results demonstrate the effectiveness of the proposed scheme in achieving high computing efficiency gain, as compared with benchmark schemes.

2401.10157

Marina Krstic Marinkovic

Sandip Maiti, Debasish Banerjee, Shailesh Chandrasekharan, Marina Krstic Marinkovic

A qubit regularization of asymptotic freedom without fine-tuning

8 pages, 8 figures, contribution to the 40th International Symposium on Lattice Field Theory (Lattice 2023), July 31st - August 4th, 2023, Fermi National Accelerator Laboratory, Batavia, Illinois, USA

hep-lat cond-mat.str-el hep-th quant-ph

http://creativecommons.org/licenses/by-nc-nd/4.0/

Other than the commonly used Wilson's regularization of quantum field theories (QFTs), there is a growing interest in regularizations that explore lattice models with a strictly finite local Hilbert space, in anticipation of the upcoming era of quantum simulations of QFTs. A notable example is Euclidean qubit regularization, which provides a natural way to recover continuum QFTs that emerge via infrared fixed points of lattice theories. Can such regularizations also capture the physics of ultraviolet fixed points? We present a novel regularization of the asymptotically free massive continuum QFT that emerges at the Berezenski-Kosterlitz-Thouless (BKT) transition through a hard core loop-gas model, discussing the advantages this model provides compared to traditional regularizations. In particular, we demonstrate that without the need for fine-tuning, it can reproduce the universal step-scaling function of the classical lattice XY model in the massive phase as we approach the phase transition.

2401.10158

Nikolaos Koursioumpas

Nikolaos Koursioumpas, Lina Magoula, Ioannis Stavrakakis, Nancy Alonistioti, M. A. Gutierrez-Estevez, Ramin Khalili

DISTINQT: A Distributed Privacy Aware Learning Framework for QoS Prediction for Future Mobile and Wireless Networks

11 Pages Double Column, 9 Figures, Submitted for possible publication in the IEEE Transactions on Vehicular Technology (IEEE TVT)

cs.NI cs.AI cs.CR cs.DC cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Beyond 5G and 6G networks are expected to support new and challenging use cases and applications that depend on a certain level of Quality of Service (QoS) to operate smoothly. Predicting the QoS in a timely manner is of high importance, especially for safety-critical applications as in the case of vehicular communications. Although until recent years the QoS prediction has been carried out by centralized Artificial Intelligence (AI) solutions, a number of privacy, computational, and operational concerns have emerged. Alternative solutions have been surfaced (e.g. Split Learning, Federated Learning), distributing AI tasks of reduced complexity across nodes, while preserving the privacy of the data. However, new challenges rise when it comes to scalable distributed learning approaches, taking into account the heterogeneous nature of future wireless networks. The current work proposes DISTINQT, a privacy-aware distributed learning framework for QoS prediction. Our framework supports multiple heterogeneous nodes, in terms of data types and model architectures, by sharing computations across them. This, enables the incorporation of diverse knowledge into a sole learning process that will enhance the robustness and generalization capabilities of the final QoS prediction model. DISTINQT also contributes to data privacy preservation by encoding any raw input data into a non-linear latent representation before any transmission. Evaluation results showcase that our framework achieves a statistically identical performance compared to its centralized version and an average performance improvement of up to 65% against six state-of-the-art centralized baseline solutions in the Tele-Operated Driving use case.

2401.10159

St\'ephane Launois

St\'ephane Launois and Tom Lenagan

Derivations and the first Hochschild cohomology group of the quantum grassmannian

30 pages; comments welcome

math.QA math.RA

http://creativecommons.org/licenses/by/4.0/

We calculate the derivations and the first Hochschild cohomology group of the quantum grassmannian over a field of characteristic zero in the generic case when the deformation parameter is not a root of unity. Using graded techniques and two special homogeneous normal elements of the quantum grassmannian, we reduce the problem to computing derivations of the quantum grassmannian that act trivially on these two normal elements. We then use the dehomogenisation equality which shows that a localisation of the quantum grassmannian is equal to a skew Laurent extension of quantum matrices. This equality is used to connect derivations of the quantum grassmannian with those of quantum matrices. More precisely, again using graded techniques, we show that derivations of the quantum grassmannian that act trivially on our two normal elements restrict to homogeneous derivations of quantum matrices. The derivations of quantum matrices are known in the square case and technical details needed to deal with the general case are given in an appendix. This allows us to explicitly describe the first Hochschild cohomology group of the quantum grassmannian.

2401.10160

Alexander Ulanov

Thibault Wildi, Alexander E. Ulanov, Thibault Voumard, Bastian Ruhnke, Tobias Herr

Phase-stabilised self-injection-locked microcomb

7 pages, 4 figures

physics.optics physics.app-ph

http://creativecommons.org/licenses/by-nc-sa/4.0/

Microresonator frequency combs (microcombs) hold potential for precision metrology in a compact form factor impacting applications such as point-of-care diagnostics, environmental monitoring, time-keeping, navigation and astronomy. Through the principle of self-injection locking, electrically-driven chip-based microcombs with low complexity are now possible. However, phase-stabilisation of such self-injection-locked microcombs, a prerequisite for metrological frequency combs, has yet to be attained. Here, addressing this critical need, we demonstrate full phase-stabilisation of a self-injection-locked microcomb. The microresonator is implemented in a silicon nitride photonic chip, and by controlling a pump laser diode and a microheater with low voltage signals (sub 1.5 V), we achieve independent actuation of the comb's offset and line spacing frequencies. Both actuators reach a bandwidth of over 100 kHz and permit phase-locking of the microcomb to external frequency references. These results establish photonic chip-based, self-injection-locked microcombs as a low-complexity, yet versatile source for coherent precision metrology in emerging applications.

2401.10161

Fernando Garc\'ia-Casta\~no

Fernando Garc\'ia-Casta\~no and M.A. Melguizo Padial

A Set-Valued Lagrange Theorem based on a Process for Convex Vector Programming

Optimization. Volume 68, Issue 11, 2019

10.1080/02331934.2019.1616731

math.OC

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this paper, we present a new set-valued Lagrange multiplier theorem for constrained convex set-valued optimization problems. We introduce the novel concept of Lagrange process. This concept is a natural extension of the classical concept of Lagrange multiplier where the conventional notion of linear continuous operator is replaced by the concept of closed convex process, its set-valued analogue. The behaviour of this new Lagrange multiplier based on a process is shown to be particularly appropriate for some types of proper minimal points and, in general, when it has a bounded base.

2401.10162

Xuan Mei

Garvit Arora, Shubhangi Shubhangi, Ying Wu, Xuan Mei

An Exploration to the Correlation Structure and Clustering of Macroeconomic Variables (MEV)

q-fin.RM

http://creativecommons.org/licenses/by/4.0/

As a quantitative characterization of the complicated economy, Macroeconomic Variables (MEVs), including GDP, inflation, unemployment, income, spending, interest rate, etc., are playing a crucial role in banks' portfolio management and stress testing exercise. In recent years, especially during the COVID-19 period and the current high inflation environment, people are frequently talking about the changing "correlation structure" of MEVs. In this paper, we use a principal component based algorithm to better understand MEVs' correlation structure in a given period. We also demonstrate how this method can be used to visualize historical MEVs pattern changes between 2000 and 2022. Further, we use this method to compare different hypothetical or historical macroeconomic scenarios and present our key findings.

2401.10163

Marco Rip\`a

Marco Rip\`a

Solving the $106$ years old $3^k$ points problem with the clockwise-algorithm

17 pages, 12 figures. A video animation of the solution from 1 to 4 dimensions can be found on YouTube (https://www.youtube.com/watch?v=SSL9R0hQRKM)

Journal of Fundamental Mathematics and Applications, Vol. 3 (2020), No. 2, pp. 84-97

10.14710/jfma.v3i2.8551

math.GM

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this paper, we present the clockwise-algorithm that solves the extension in $k$-dimensions of the infamous nine-dot problem, the well-known two-dimensional thinking outside the box puzzle. We describe a general strategy that constructively produces minimum length covering trails, for any $k \in \mathbb{N}-\{0\}$, solving the NP-complete $(3 \times 3 \times \cdots \times 3)$-point problem inside $3 \times 3 \times \cdots \times 3$ hypercubes. In particular, using our algorithm, we explicitly draw different covering trails of minimal length $h(k)=\frac{3^k-1}{2}$, for $k=3, 4, 5$. Furthermore, we conjecture that, for every $k \geq 1$, it is possible to solve the $3^k$-point problem with $h(k)$ lines starting from any of the $3^k$ nodes, except from the central one. Finally, we cover a $3 \times 3 \times 3$ grid with a tree of size $12$.

2401.10164

Sonia G\'amiz-Fortis

Matilde Garc\'ia-Valdecasas Ojeda, Patricio Yeste, Sonia Raquel G\'amiz-Fortis, Yolanda Castro-D\'iez, and Mar\'ia Jes\'us Esteban-Parra

Future changes in land and atmospheric variables: An analysis of their couplings in the Iberian Peninsula

Science of the Total Environment, Volume 72220, June 2020, Article number 137902

10.1016/j.scitotenv.2020.137902

physics.ao-ph

http://creativecommons.org/licenses/by-nc-nd/4.0/

This work investigates climate--projections over a transitional region between dry and wet climates, the Iberian Peninsula (IP). With this purpose, the Weather Research and Forecasting (WRF) model, driven CCSM4 and MPI-ESM-LR GCMs previously bias-corrected, was used. Simulations were carried out for two periods, 1980-2014 and 2071-2100, and under RCP4.5 and RCP8.5. The analysis focused on changes in land-surface processes, their causes, and the potential impact on the climate system. To achieve this, seasonal projected changes of land-surface (soil moisture and surface evapotranspiration) and atmospheric variables involved in the hydrologic (precipitation and runoff) and energy balance (temperature and solar incoming radiation) were investigated. The results reveal that the IP is likely to experience a soil dryness by the end of the 21st century, particularly during summer and fall, more apparent in the southern IP, and stronger under the RCP8.5. However, such trends would have different implications throughout the year and directly affect the surface evapotranspiration. Moreover, soil-drying trends are mainly associated with reductions in the large-scale precipitation during spring, summer, and fall and by enhanced evapotranspiration particularly in spring over the northwestern IP. In addition, the results show notably changes in soil conditions at high altitude, particularly during winter, which may alter the land-atmosphere processes that currently occur in these regions. In this context, noteworthy changes in the climate system are expected, leading to adverse impacts on water resources and temperature. The results highlight the complex and nonlinear nature of land-atmosphere interactions in regions such as the IP, which is a tremendous challenge for adequately developing mitigation and adaptation strategies to anthropogenic climate change.

2401.10165

Francesco Benini

Andrea Antinucci, Francesco Benini

Anomalies and gauging of U(1) symmetries

Two columns, 11 pages

hep-th

http://creativecommons.org/licenses/by/4.0/

We propose the Symmetry TFT for theories with a $U(1)$ symmetry in arbitrary dimension. The Symmetry TFT describes the structure of the symmetry, its anomalies, and the possible topological manipulations. It is constructed as a BF theory of gauge fields for groups $U(1)$ and $\mathbb{R}$, and contains a continuum of topological operators. We also propose an operation that produces the Symmetry TFT for the theory obtained by dynamically gauging the $U(1)$ symmetry. We discuss many examples. As an interesting outcome, we obtain the Symmetry TFT for the non-invertible $\mathbb{Q}/\mathbb{Z}$ chiral symmetry in four dimensions.

2401.10166

Yunjie Tian

Yue Liu and Yunjie Tian and Yuzhong Zhao and Hongtian Yu and Lingxi Xie and Yaowei Wang and Qixiang Ye and Yunfan Liu

VMamba: Visual State Space Model

13 pages, 6 figures, 4 tables

cs.CV

http://creativecommons.org/licenses/by-sa/4.0/

Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) stand as the two most popular foundation models for visual representation learning. While CNNs exhibit remarkable scalability with linear complexity w.r.t. image resolution, ViTs surpass them in fitting capabilities despite contending with quadratic complexity. A closer inspection reveals that ViTs achieve superior visual modeling performance through the incorporation of global receptive fields and dynamic weights. This observation motivates us to propose a novel architecture that inherits these components while enhancing computational efficiency. To this end, we draw inspiration from the recently introduced state space model and propose the Visual State Space Model (VMamba), which achieves linear complexity without sacrificing global receptive fields. To address the encountered direction-sensitive issue, we introduce the Cross-Scan Module (CSM) to traverse the spatial domain and convert any non-causal visual image into order patch sequences. Extensive experimental results substantiate that VMamba not only demonstrates promising capabilities across various visual perception tasks, but also exhibits more pronounced advantages over established benchmarks as the image resolution increases. Source code has been available at https://github.com/MzeroMiko/VMamba.

2401.10167

Emily Pass

Emily K. Pass, David Charbonneau, David W. Latham, Perry Berlind, Michael L. Calkins, Gilbert A. Esquerdo, Jessica Mink

The Mass Dependence of H{\alpha} Emission and Stellar Spindown for Fully Convective M Dwarfs

Submitted to ApJ; 18 pages, 8 figures, 5 tables

astro-ph.SR astro-ph.EP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Fully convective M dwarfs typically remain rapidly rotating and magnetically active for billions of years, followed by an abrupt and mass-dependent transition to slow rotation and quiescence. A robust understanding of this process is complicated by difficulties in estimating M-dwarf ages and potential dependencies on other variables such as birth environment or metallicity. To isolate the effect of mass, we consider M dwarfs in wide binaries. We identify 67 widely separated, fully convective (0.08-0.35M$_\odot$) M-dwarf binary systems using Gaia and measure the H$\alpha$ feature for each component. We classify the pairs into three categories: systems where both components are active, systems where both are inactive, and candidate transition systems, where one component is active and the other inactive. We gather higher-resolution spectra of the candidate transition systems to verify that their behavior does not result from an unresolved third component, yielding one new triple with surprising activity levels. Neglecting this triple, we find 22 active, 36 inactive, and 8 transition pairs. Our results are consistent with the epoch of spindown for these binaries being primarily determined by mass, with mild second-order effects; we place a 1$\sigma$ upper limit of 0.5Gyr or 25% on the dispersion in the mass-dependent spindown relation. Our findings suggest that the large dispersion in spindown epoch previously observed for field stars of a given mass may stem from differences in birth environment, in addition to modest intrinsic stochasticity. We also see evidence that the wide binary population is dispersed over time due to dynamical processing.

2401.10168

Keren Sharon

Maxwell Klein, Keren Sharon, Kate Napier, Michael D. Gladders, Gourav Khullar, Matthew Bayliss, H{\aa}kon Dahle, M. Riley Owens, Antony Stark, Sasha Brownsberger, Keunho J. Kim, Nicole Kuchta, Guillaume Mahler, Grace Smith, Ryan Walker, Katya Gozman, Michael N. Martinez, Owen S. Matthews Acu\~na, Kaiya Merz, Jorge A. Sanchez, Daniel J. Kavin Stein, Ezra O. Sukay, Kiyan Tavangar

COOL-LAMPS VI: Lens model and New Constraints on the Properties of COOL J1241+2219, a Bright z = 5 Lyman Break Galaxy and its z = 1 Cluster Lens

Submitted to ApJ

astro-ph.GA

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We present a strong lensing analysis of COOL J1241+2219, the brightest known gravitationally lensed galaxy at $z \geq 5$, based on new multi-band Hubble Space Telescope (HST) imaging data. The lensed galaxy has a redshift of z=5.043, placing it shortly after the end of the Epoch of Reionization, and an AB magnitude z_AB=20.47 mag (Khullar et al. 2021). As such, it serves as a touchstone for future research of that epoch. The high spatial resolution of HST reveals internal structure in the giant arc, from which we identify 15 constraints and construct a robust lens model. We use the lens model to extract cluster mass and lensing magnification. We find that the mass enclosed within the Einstein radius of the z=1.001 cluster lens is M(<5.77'')=$1.079^{+0.023}_{-0.007}$, significantly lower than other known strong lensing clusters at its redshift. The average magnification of the giant arc is $<\mu_{arc}>=76^{+40}_{-20}$, a factor of $2.4^{+1.4}_{-0.7}$ greater than previously estimated from ground-based data; the flux-weighted average magnification is $<\mu_{arc}>=92^{+37}_{-31}$ We update the current measurements of the stellar mass and star formation rate (SFR) of the source for the revised magnification, $\log(M_\star/M_{\odot})=9.7\pm0.3$ and ${\rm SFR} = 10.3^{+7.0}_{-4.4}$ $ M_{\odot} $yr$^{-1}$. The powerful lensing magnification acting upon COOL J1241+2219 resolves the source and enables future studies of the properties of its star formation on a clump-by-clump basis. The lensing analysis presented here will support upcoming multiwavelength characterization with HST and JWST data of the stellar mass assembly and physical properties of this high-redshift lensed galaxy.

2401.10169

Ales Wodecki

Ale\v{s} Wodecki

On Taylor-like Estimates for $L^{2}$ Polynomial Approximations

10 pages, 0 figures

math.GM

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Polynomial series approximations are a central theme in approximation theory. Two types of series, which are featured most prominently in pure and applied mathematics, are Taylor series expansions and expansions derived based on families of $L^{2}-$orthogonal polynomials on bounded intervals. The identity theorem implies that all such series agree on $\mathbb{C}$ in the limit. This motivates an effort to derive properties, which relate the original function to a truncated series based on $L^{2}-$orthogonal polynomial approximation which hold on unbounded intervals. In particular, the Chebyshev series expansion of $e^{x}$ is studied and an algebraic criterion which can be used to confirm bounds analogous to the Taylor series upper and lower bound estimates for $x<0$ is provided. In addition to this, the generalizations to other settings involving different functions and alternative families of $L^{2}$-orthogonal polynomials are discussed.

2401.10170

Cedric Fraces

Cedric Fraces Gasmi, Wennan Long

Building a Life Cycle Assessment Model using Bayesian Networks

physics.data-an

http://creativecommons.org/licenses/by/4.0/

This paper introduces the Oilfield Pollutant Graphical Model (OPGM), an innovative approach designed to improve the benchmarking and uncertainty analysis of greenhouse gas (GHG) emissions in oilfields. Building on the robust foundation provided by the Oil Production Greenhouse Gas Emission Estimator (OPGEE) framework, OPGM retains all essential functionalities of the latest OPGEE iteration (v3.0c), while offering substantial improvements in user experience and computational performance. Key advances of OPGM include a streamlined user interface for more intuitive interaction, which facilitates transparent visualization of intermediate results and thus contributes to a more interpretable and accessible analysis process. A notable feature of the OPGM is its ability to naturally perform sensitivity analyzes. This is achieved by allowing users to seamlessly transition nodes from deterministic to probabilistic, thereby integrating uncertainty directly into the core structure of the model. OPGM achieves remarkable computational efficiency, executing analyzes at a speed 1e+5 times faster than the Excel-based OPGEE, thus facilitating rapid large-scale emissions assessments. This leap in processing speed represents a significant step forward in emissions modeling, enabling more agile and accurate environmental impact assessments. The integration of OPGM into existing Life Cycle Assessment (LCA) practices holds the promise of significantly improving the precision and speed of environmental impact analyses, offering a vital tool for policymakers and industry stakeholders in their efforts to better understand and manage the environmental impacts of oilfield operations.

2401.10171

Andreas Engelhardt

Andreas Engelhardt, Amit Raj, Mark Boss, Yunzhi Zhang, Abhishek Kar, Yuanzhen Li, Deqing Sun, Ricardo Martin Brualla, Jonathan T. Barron, Hendrik P. A. Lensch, Varun Jampani

SHINOBI: Shape and Illumination using Neural Object Decomposition via BRDF Optimization In-the-wild

cs.CV cs.GR

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We present SHINOBI, an end-to-end framework for the reconstruction of shape, material, and illumination from object images captured with varying lighting, pose, and background. Inverse rendering of an object based on unconstrained image collections is a long-standing challenge in computer vision and graphics and requires a joint optimization over shape, radiance, and pose. We show that an implicit shape representation based on a multi-resolution hash encoding enables faster and robust shape reconstruction with joint camera alignment optimization that outperforms prior work. Further, to enable the editing of illumination and object reflectance (i.e. material) we jointly optimize BRDF and illumination together with the object's shape. Our method is class-agnostic and works on in-the-wild image collections of objects to produce relightable 3D assets for several use cases such as AR/VR, movies, games, etc. Project page: https://shinobi.aengelhardt.com Video: https://www.youtube.com/watch?v=iFENQ6AcYd8&feature=youtu.be

2401.10172

Geoff Vooys

Geoff Vooys

Categories of Pseudocones and Equivariant Descent

arXiv admin note: text overlap with arXiv:2110.01130

math.AG math.AT math.CT

http://creativecommons.org/licenses/by/4.0/

In this monograph we provide an in-depth and systematic study of pseudolimits of pseudofunctors $F:\mathscr{C}^{op} \to \mathfrak{Cat}$ in the $2$-category of categories where $\mathscr{C}$ is a $1$-category and use this to give an explicit and careful study of the category theory used in representation theory, equivariant algebraic geometry, and equivariant algebraic topology and give a unifying language to study equivariant sheaves, equivariant perverse sheaves, and their equivariant derived categories. We show how to use the pseudocone construction $\mathsf{Bicat}(\mathscr{C}^{op},\mathfrak{Cat})(\operatorname{cnst}(1),F)$ in order to derive categorical and homological properties of the pseudolimit of $F$. We explicitly show when the pseudolimit of $F$ is complete, cocomplete, enriched in models of a Lawvere theory, (braided) monoidal, regular, triangulated, admits $t$-structures, and more. We use these various structural results to give a new category-theoretic proof and construction of the equivariant standard and pervese $t$-structures and equivariant six functor formalism for the equivariant derived category $D_G^b(X)$ in both the geometric and topological cases as well as for $D_G^b(X;\overline{\mathbb{Q}}_{\ell})$ in the geometric case. We also show in what sense precise sense we can view the equivariant derived category in terms of localizations. After restricting to the case of group resolution categories, we show the existence of a natural isomorphism $\Theta:\alpha_X^{\ast} \Rightarrow \pi_2^{\ast}$ which satisfies a pseudofunctorial version of the cocycle condition $d_1^{\ast}\Theta = d_2^{\ast}\Theta \circ d_0^{\ast}\Theta$. We also use the pseudocone formalism to give an in-depth analysis of change of groups functors. We use the pseudocone formalism and $\Theta$ to develop a notion of equivariant trace with an eye towards the representation theory of $p$-adic groups.

2401.10173

Aaron Alejo

J. Pe\~nas, A. Bembibre, D. Cortina-Gil, L. Martin, A. Reija, C. Ruiz, M. Seimetz, A. Alejo, J. Benlliure

A multi-shot target-wheel assembly for high-repetition-rate, laser-driven proton acceleration

8 pages, 5 figures

physics.plasm-ph physics.acc-ph

http://creativecommons.org/licenses/by/4.0/

A multi-shot target assembly and automatic alignment procedure for laser-plasma proton acceleration at high-repetition-rate are introduced. The assembly is based on a multi-target rotating wheel capable of hosting $>$5000 targets, mounted on a three-dimensional motorised stage to allow rapid replenishment and alignment of the target material between laser irradiations. The automatic alignment procedure consists of a detailed mapping of the impact positions at the target surface prior to the irradiation that ensures stable operation of the target, which alongside the purpose-built design of the target wheel, enable the operation at rates up to 10 Hz. Stable and continuous laser-driven proton acceleration is demonstrated, with observed cut-off energy stability about 15%.

2401.10174

Geoff Vooys

Geoff Vooys

On the Equivariant Derived Category of Perverse Sheaves

40 Pages. Comments welcome!

math.AG math.CT

http://creativecommons.org/licenses/by/4.0/

In this paper we extend Beilinson's realization formalism for triangulated categories and filtered triangulated categories to a pseudofunctorial and pseudonatural setting. As a consequence we prove an equivariant version of Beilinson's Theorem: for any algebraic group $G$ over an algebraically closed field $K$ and for any $G$-variety $X$, there is an equivalence of categories $D_G^b(X; \overline{\mathbb{Q}}_{\ell}) \simeq D_G^b(\mathbf{Perv}(X;\overline{\mathbb{Q}}_{\ell}))$ where $\ell$ is an integer prime coprime to the characteristic of $K$. We also show that the equivariant analogues of the other non-$D$-module aspects of Beilinson's Theorem hold in the equivariant case.

2401.10175

Zhaobo Zheng

Zhaobo Zheng, Kumar Akash, and Teruhisa Misu

DualTake: Predicting Takeovers across Mobilities for Future Personalized Mobility Services

4 pages, 6 figures. Accepted to IEEE Human Robot Interaction (HRI)

10.1145/3610978.3640610

cs.HC

http://creativecommons.org/licenses/by-nc-nd/4.0/

A hybrid society is expected to emerge in the near future, with different mobilities interacting together, including cars, micro-mobilities, pedestrians, and robots. People may utilize multiple types of mobilities in their daily lives. As vehicle automation advances, driver modeling flourishes to provide personalized intelligent services. Thus, modeling drivers across mobilities would pave the road for future society mobility-as-a-service, and it is particularly interesting to predict driver behaviors in newer mobilities with traditional mobility data. In this work, we present takeover prediction on a micro-mobility, with car simulation data.The promising model performance demonstrates the feasibility of driver modeling across mobilities, as the first in the field.

2401.10176

Amol Khanna

Anish Lakkapragada, Amol Khanna, Edward Raff, Nathan Inkawhich

Comprehensive OOD Detection Improvements

cs.LG cs.CV

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

As machine learning becomes increasingly prevalent in impactful decisions, recognizing when inference data is outside the model's expected input distribution is paramount for giving context to predictions. Out-of-distribution (OOD) detection methods have been created for this task. Such methods can be split into representation-based or logit-based methods from whether they respectively utilize the model's embeddings or predictions for OOD detection. In contrast to most papers which solely focus on one such group, we address both. We employ dimensionality reduction on feature embeddings in representation-based methods for both time speedups and improved performance. Additionally, we propose DICE-COL, a modification of the popular logit-based method Directed Sparsification (DICE) that resolves an unnoticed flaw. We demonstrate the effectiveness of our methods on the OpenOODv1.5 benchmark framework, where they significantly improve performance and set state-of-the-art results.

2401.10177

Alejandro Mus

A. Mus, I. Mart\'i-Vidal

New-generation Maximum Entropy Method (ngMEM): a Lagrangian-based algorithm for dynamic reconstruction of interferometric data

astro-ph.IM

http://creativecommons.org/licenses/by/4.0/

Imaging interferometric data in radio astronomy requires the use of non-linear algorithms that rely on different assumptions on the source structure and may produce non-unique results. This is especially true for Very Long Baseline Interferometry (VLBI) observations, where the sampling of Fourier space is very sparse. A basic tenet in standard VLBI imaging techniques is to assume that the observed source structure does not evolve during the observation. However, the recent VLBI results of the supermassive black hole (SMBH) at our Galactic Center (Sagittarius A$^*$, SgrA*), recently reported by the Event Horizon Telescope Collaboration (EHTC), require the development of dynamic imaging algorithms, since it exhibits variability at minute timescales. In this paper, we introduce a new non-convex optimization problem that extends the standard Maximum Entropy Method (MEM), for reconstructing intra-observation dynamical images from interferometric data that evolves in every integration time. We present a rigorous mathematical formalism to solve the problem via the primal-dual approach. We build a Newton strategy and we give its numerical complexity. We also give a strategy to iteratively improve the obtained solution and finally, we define a novel figure of merit to evaluate the quality of the recovered solution. Then, we test the algorithm, called ngMEM, in different synthetic datasets, with increasing difficulty. Finally, we compare it with another well-established dynamical imaging method. Within this comparison we identified a significant improvement of the ngMEM reconstructions. Moreover, the evaluation of the integration time evolution scheme and the time contribution showed to play a crucial role for obtaining good dynamic reconstructions.

2401.10178

Zahra Babaiee

Zahra Babaiee, Peyman M. Kiasari, Daniela Rus, Radu Grosu

Neural Echos: Depthwise Convolutional Filters Replicate Biological Receptive Fields

Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (2024) 8216-8225

cs.CV cs.AI cs.NE

http://creativecommons.org/licenses/by-nc-sa/4.0/

In this study, we present evidence suggesting that depthwise convolutional kernels are effectively replicating the structural intricacies of the biological receptive fields observed in the mammalian retina. We provide analytics of trained kernels from various state-of-the-art models substantiating this evidence. Inspired by this intriguing discovery, we propose an initialization scheme that draws inspiration from the biological receptive fields. Experimental analysis of the ImageNet dataset with multiple CNN architectures featuring depthwise convolutions reveals a marked enhancement in the accuracy of the learned model when initialized with biologically derived weights. This underlies the potential for biologically inspired computational models to further our understanding of vision processing systems and to improve the efficacy of convolutional networks.

2401.10179

Siva Athreya

Siva Athreya, Alexander Drewitz and Rongfeng Sun

An Invariance Principle for a Random Walk Among Moving Traps via Thermodynamic Formalism

41 pages

math.PR math.DS

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We consider a random walk among a Poisson cloud of moving traps on ${\mathbb Z}^d$, where the walk is killed at a rate proportional to the number of traps occupying the same position. In dimension $d=1$, we have previously shown that under the annealed law of the random walk conditioned on survival up to time $t$, the walk is sub-diffusive. Here we show that in $d\geq 6$ and under diffusive scaling, this annealed law satisfies an invariance principle with a positive diffusion constant if the killing rate is small. Our proof is based on the theory of thermodynamic formalism, where we extend some classic results for Markov shifts with a finite alphabet and a potential of summable variation to the case of an uncountable non-compact alphabet.

2401.10180

Javier Enrique Aguilar Romero

Javier Enrique Aguilar, Paul-Christian B\"urkner

Generalized Decomposition Priors on R2

31 pages, 12 figures

stat.ME

http://creativecommons.org/licenses/by/4.0/

The adoption of continuous shrinkage priors in high-dimensional linear models has gained momentum, driven by their theoretical and practical advantages. One of these shrinkage priors is the R2D2 prior, which comes with intuitive hyperparameters and well understood theoretical properties. The core idea is to specify a prior on the percentage of explained variance $R^2$ and to conduct a Dirichlet decomposition to distribute the explained variance among all the regression terms of the model. Due to the properties of the Dirichlet distribution, the competition among variance components tends to gravitate towards negative dependence structures, fully determined by the individual components' means. Yet, in reality, specific coefficients or groups may compete differently for the total variability than the Dirichlet would allow for. In this work we address this limitation by proposing a generalization of the R2D2 prior, which we term the Generalized Decomposition R2 (GDR2) prior. Our new prior provides great flexibility in expressing dependency structures as well as enhanced shrinkage properties. Specifically, we explore the capabilities of variance decomposition via logistic normal distributions. Through extensive simulations and real-world case studies, we demonstrate that GDR2 priors yield strongly improved out-of-sample predictive performance and parameter recovery compared to R2D2 priors with similar hyper-parameter choices.

2401.10181

Amine Ouazad

Amine C-L. Ouazad

Equilibrium Multiplicity: A Systematic Approach using Homotopies, with an Application to Chicago

econ.GN q-fin.EC

http://creativecommons.org/licenses/by/4.0/

Discrete choice models with social interactions or spillovers may exhibit multiple equilibria. This paper provides a systematic approach to enumerating them for a quantitative spatial model with discrete locations, social interactions, and elastic housing supply. The approach relies on two homotopies. A homotopy is a smooth function that transforms the solutions of a simpler city where solutions are known, to a city with heterogeneous locations and finite supply elasticity. The first homotopy is that, in the set of cities with perfectly elastic floor surface supply, an economy with heterogeneous locations is homotopic to an economy with homogeneous locations, whose solutions can be comprehensively enumerated. Such an economy is epsilon close to an economy whose equilibria are the zeros of a system of polynomials. This is a well-studied area of mathematics where the enumeration of equilibria can be guaranteed. The second homotopy is that a city with perfectly elastic housing supply is homotopic to a city with an arbitrary supply elasticity. In a small number of cases, the path may bifurcate and a single path yields two or more equilibria. By running the method on thousands of cities, we obtain a large number of equilibria. Each equilibrium has different population distributions. We provide a method that is computationally feasible for economies with a large number of locations choices, with an empirical application to the City of Chicago. There exist multiple ``counterfactual Chicagos'' consistent with the estimated parameters. Population distribution, prices, and welfare are not uniquely pinned down by amenities. The paper's method can be applied to models in trade and IO. Further applications of algebraic geometry are suggested.

2401.10182

Kanad Sengupta

Kanad Sengupta, K. Muhammed Shafi, Soumya Asokan, C. M. Chandrashekar

Experimental demonstration of quantum illumination using polarization-entangled photon pairs and CHSH value as measure

9 Pages, 8 Figures

quant-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Entangled light sources for illuminating objects offers advantages over conventional illumination methods by enhancing the detection sensitivity of a reflecting object. The crux of the quantum advantage lies in way we can practically leverage quantum correlations to isolate the background noise and detect the low reflectivity object. In this work we experimentally demonstrated the advantages of using polarization-entangled photon pairs for quantum illumination and show that the quantum correlation measure using CHSH value is robust against background noise and losses. We also show that the residual of quantum correlations help in identifying the object of reflectivity, $\eta$ as low as 0.05 and when signal-to-noise ratio is as low as 0.003 for $\eta=0.7$, surpassing the earlier demonstrated results. Robustness of correlation measure with photon attenuation in atmospheric condition is analysed to show the practical feasibility of the real time application.

2401.10183

Ariel Weiss

Amit Ophir, Ariel Weiss

Maximal stable lattices in representations over discretely valued fields

13 pages, comments welcome!

math.NT math.GR math.RT

http://creativecommons.org/licenses/by/4.0/

Let $\rho\colon G\to \mathrm{GL}_n(K)$ be an continuous irreducible representation of a compact group over a complete discretely valued field $K$. Let $W_i,W_j$ be two irreducible subrepresentations of $\overline{\rho}^{ss}$, the semisimplification of the residual representation. We study the structure of the $G$-stable lattices $\Lambda\subseteq K^n$ with a view to understanding the question of when $\rho$ realises a non-split extension of $W_i$ by $W_j$. In particular, we introduce the notion of a maximal $G$-stable lattice and prove that any non-split extension of $W_i$ by $W_j$ that can be realised by $\rho$ can also be realised by a maximal lattice. As applications, we give a new proof and a strengthening of Bella\"iche's generalisation of Ribet's Lemma, which assures the abundancy of non-split extensions that can be realised by $\rho$. On the other hand, we also show that, if the representations $W_i, W_j$ occur with multiplicity one in $\overline{\rho}^{ss}$, then $\rho$ can realise at most one non-split extension of $W_i$ by $W_j$.

2401.10184

Albatool Wazzan

Albatool Wazzan, Stephen MacNeil, Richard Souvenir

Comparing Traditional and LLM-based Search for Image Geolocation

10.1145/3627508.3638305

cs.IR cs.HC

http://creativecommons.org/licenses/by-sa/4.0/

Web search engines have long served as indispensable tools for information retrieval; user behavior and query formulation strategies have been well studied. The introduction of search engines powered by large language models (LLMs) suggested more conversational search and new types of query strategies. In this paper, we compare traditional and LLM-based search for the task of image geolocation, i.e., determining the location where an image was captured. Our work examines user interactions, with a particular focus on query formulation strategies. In our study, 60 participants were assigned either traditional or LLM-based search engines as assistants for geolocation. Participants using traditional search more accurately predicted the location of the image compared to those using the LLM-based search. Distinct strategies emerged between users depending on the type of assistant. Participants using the LLM-based search issued longer, more natural language queries, but had shorter search sessions. When reformulating their search queries, traditional search participants tended to add more terms to their initial queries, whereas participants using the LLM-based search consistently rephrased their initial queries.

2401.10185

Sourish Gunesh Dhekane

Sourish Gunesh Dhekane, Thomas Ploetz

Transfer Learning in Human Activity Recognition: A Survey

40 pages, 5 figures, 7 tables

cs.LG eess.SP

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sensor-based human activity recognition (HAR) has been an active research area, owing to its applications in smart environments, assisted living, fitness, healthcare, etc. Recently, deep learning based end-to-end training has resulted in state-of-the-art performance in domains such as computer vision and natural language, where large amounts of annotated data are available. However, large quantities of annotated data are not available for sensor-based HAR. Moreover, the real-world settings on which the HAR is performed differ in terms of sensor modalities, classification tasks, and target users. To address this problem, transfer learning has been employed extensively. In this survey, we focus on these transfer learning methods in the application domains of smart home and wearables-based HAR. In particular, we provide a problem-solution perspective by categorizing and presenting the works in terms of their contributions and the challenges they address. We also present an updated view of the state-of-the-art for both application domains. Based on our analysis of 205 papers, we highlight the gaps in the literature and provide a roadmap for addressing them. This survey provides a reference to the HAR community, by summarizing the existing works and providing a promising research agenda.

2401.10186

Zden\v{e}k Kasner

Zden\v{e}k Kasner, Ond\v{r}ej Du\v{s}ek

Beyond Reference-Based Metrics: Analyzing Behaviors of Open LLMs on Data-to-Text Generation

26 pages

cs.CL

http://creativecommons.org/licenses/by-nc-sa/4.0/

We investigate to which extent open large language models (LLMs) can generate coherent and relevant text from structured data. To prevent bias from benchmarks leaked into LLM training data, we collect Quintd-1: an ad-hoc benchmark for five data-to-text (D2T) generation tasks, consisting of structured data records in standard formats gathered from public APIs. We leverage reference-free evaluation metrics and LLMs' in-context learning capabilities, allowing us to test the models with no human-written references. Our evaluation focuses on annotating semantic accuracy errors on token-level, combining human annotators and a metric based on GPT-4. Our systematic examination of the models' behavior across domains and tasks suggests that state-of-the-art open LLMs with 7B parameters can generate fluent and coherent text from various standard data formats in zero-shot settings. However, we also show that semantic accuracy of the outputs remains a major issue: on our benchmark, 80% of outputs of open LLMs contain a semantic error according to human annotators (91% according to GPT-4). Our code, data, and model outputs are available at https://d2t-llm.github.io.

2401.10187

Abhinav Jangda

Abhinav Jangda and Mohit Yadav

Fast Kronecker Matrix-Matrix Multiplication on GPUs

10.1145/3627535.3638489

cs.DC

http://creativecommons.org/licenses/by/4.0/

Kronecker Matrix-Matrix Multiplication (Kron-Matmul) is the multiplication of a matrix with the Kronecker Product of several smaller matrices. Kron-Matmul is a core operation for many scientific and machine learning computations. State-of-the-art Kron-Matmul implementations utilize existing tensor algebra operations, such as matrix multiplication, transpose, and tensor matrix multiplication. However, this design choice prevents several Kron-Matmul specific optimizations, thus, leaving significant performance on the table. To address this issue, we present FastKron, an efficient technique for Kron-Matmul on single and multiple GPUs. FastKron is independent of linear algebra operations enabling several new optimizations for Kron-Matmul. Thus, it performs up to 40.7x and 7.85x faster than existing implementations on 1 and 16 GPUs respectively.

2401.10188

Swarup Bhowmik

Swarup Bhowmik

Center and torsion of a quotient of the group of piecewise linear homeomorphisms of the real line

8 pages

math.GT

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

In this article, we show that the group comprising piecewise-linear homeomorphisms of the real line with bounded slopes is not simple. Furthermore, we establish that a quotient of this group is torsion-free, and importantly, the center of that quotient group is indeed trivial.

2401.10189

Qingyun Wang

Qingyun Wang, Zixuan Zhang, Hongxiang Li, Xuan Liu, Jiawei Han, Heng Ji, Huimin Zhao

Chem-FINESE: Validating Fine-Grained Few-shot Entity Extraction through Text Reconstruction

16 pages. Accepted by Findings of the Association for Computational Linguistics: EACL 2024. Code and resources are available at https://github.com/EagleW/Chem-FINESE

cs.CL cs.AI cs.LG

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Fine-grained few-shot entity extraction in the chemical domain faces two unique challenges. First, compared with entity extraction tasks in the general domain, sentences from chemical papers usually contain more entities. Moreover, entity extraction models usually have difficulty extracting entities of long-tailed types. In this paper, we propose Chem-FINESE, a novel sequence-to-sequence (seq2seq) based few-shot entity extraction approach, to address these two challenges. Our Chem-FINESE has two components: a seq2seq entity extractor to extract named entities from the input sentence and a seq2seq self-validation module to reconstruct the original input sentence from extracted entities. Inspired by the fact that a good entity extraction system needs to extract entities faithfully, our new self-validation module leverages entity extraction results to reconstruct the original input sentence. Besides, we design a new contrastive loss to reduce excessive copying during the extraction process. Finally, we release ChemNER+, a new fine-grained chemical entity extraction dataset that is annotated by domain experts with the ChemNER schema. Experiments in few-shot settings with both ChemNER+ and CHEMET datasets show that our newly proposed framework has contributed up to 8.26% and 6.84% absolute F1-score gains respectively.

2401.10190

Gil Goldshlager

Gil Goldshlager, Nilin Abrahamsen, Lin Lin

A Kaczmarz-inspired approach to accelerate the optimization of neural network wavefunctions

physics.comp-ph cs.LG physics.chem-ph

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Neural network wavefunctions optimized using the variational Monte Carlo method have been shown to produce highly accurate results for the electronic structure of atoms and small molecules, but the high cost of optimizing such wavefunctions prevents their application to larger systems. We propose the Subsampled Projected-Increment Natural Gradient Descent (SPRING) optimizer to reduce this bottleneck. SPRING combines ideas from the recently introduced minimum-step stochastic reconfiguration optimizer (MinSR) and the classical randomized Kaczmarz method for solving linear least-squares problems. We demonstrate that SPRING outperforms both MinSR and the popular Kronecker-Factored Approximate Curvature method (KFAC) across a number of small atoms and molecules, given that the learning rates of all methods are optimally tuned. For example, on the oxygen atom, SPRING attains chemical accuracy after forty thousand training iterations, whereas both MinSR and KFAC fail to do so even after one hundred thousand iterations.

2401.10191

Grzegorz Rype\'s\'c

Grzegorz Rype\'s\'c, Sebastian Cygert, Valeriya Khan, Tomasz Trzci\'nski, Bartosz Zieli\'nski, Bart{\l}omiej Twardowski

Divide and not forget: Ensemble of selectively trained experts in Continual Learning

Accepted to ICLR2024 (main track), code is available at: https://github.com/grypesc/SEED

cs.LG cs.CV

http://creativecommons.org/licenses/by/4.0/

Class-incremental learning is becoming more popular as it helps models widen their applicability while not forgetting what they already know. A trend in this area is to use a mixture-of-expert technique, where different models work together to solve the task. However, the experts are usually trained all at once using whole task data, which makes them all prone to forgetting and increasing computational burden. To address this limitation, we introduce a novel approach named SEED. SEED selects only one, the most optimal expert for a considered task, and uses data from this task to fine-tune only this expert. For this purpose, each expert represents each class with a Gaussian distribution, and the optimal expert is selected based on the similarity of those distributions. Consequently, SEED increases diversity and heterogeneity within the experts while maintaining the high stability of this ensemble method. The extensive experiments demonstrate that SEED achieves state-of-the-art performance in exemplar-free settings across various scenarios, showing the potential of expert diversification through data in continual learning.

2401.10192

Jingyi Wang

Jingyi Wang, Erin Kara, Javier A. Garc\'ia, Diego Altamirano, Tomaso Belloni, James F. Steiner, Michiel van der Klis, Adam Ingram, Guglielmo Mastroserio, Riley Connors, Matteo Lucchini, Thomas Dauser, Joseph Neilsen, Collin Lewin, Ron A. Remillard, Jeroen Homan

The 2022 Outburst of IGR J17091-3624: Connecting the exotic GRS 1915+105 to standard black hole X-ray binaries

23 pages, 15 figures, 2 tables, accepted to be published in ApJ

astro-ph.HE

http://creativecommons.org/licenses/by/4.0/

While the standard X-ray variability of black hole X-ray binaries (BHXBs) is stochastic and noisy, there are two known BHXBs that exhibit exotic `heartbeat'-like variability in their light curves: GRS 1915+105 and IGR J17091-3624. In 2022, IGR J17091-3624 went into outburst for the first time in the NICER/NuSTAR era. These exquisite data allow us to simultaneously track the exotic variability and the corresponding spectral features with unprecedented detail. We find that as in typical BHXBs, the outburst began in the hard state, then the intermediate state, but then transitioned to an exotic soft state where we identify two types of heartbeat-like variability (Class V and a new Class X). The flux-energy spectra show a broad iron emission line due to relativistic reflection when there is no exotic variability, and absorption features from highly ionized iron when the source exhibits exotic variability. Whether absorption lines from highly ionized iron are detected in IGR J17091-3624 is not determined by the spectral state alone, but rather is determined by the presence of exotic variability; in a soft spectral state, absorption lines are only detected along with exotic variability. Our finding indicates that IGR J17091-3624 can be seen as a bridge between the most peculiar BHXB GRS 1915+105 and `normal' BHXBs because it alternates between the conventional and exotic behavior of BHXBs. We discuss the physical nature of the absorbing material and exotic variability in light of this new legacy dataset.

2401.10193

James T Thorson

James T. Thorson, Sean C. Anderson, Pamela Goddard, and Christopher N. Rooper

tinyVAST: R package with an expressive interface to specify lagged and simultaneous effects in multivariate spatio-temporal models

stat.ME stat.AP

http://creativecommons.org/licenses/by-nc-sa/4.0/

Multivariate spatio-temporal models are widely applicable, but specifying their structure is complicated and may inhibit wider use. We introduce the R package tinyVAST from two viewpoints: the software user and the statistician. From the user viewpoint, tinyVAST adapts a widely used formula interface to specify generalized additive models, and combines this with arguments to specify spatial and spatio-temporal interactions among variables. These interactions are specified using arrow notation (from structural equation models), or an extended arrow-and-lag notation that allows simultaneous, lagged, and recursive dependencies among variables over time. The user also specifies a spatial domain for areal (gridded), continuous (point-count), or stream-network data. From the statistician viewpoint, tinyVAST constructs sparse precision matrices representing multivariate spatio-temporal variation, and parameters are estimated by specifying a generalized linear mixed model (GLMM). This expressive interface encompasses vector autoregressive, empirical orthogonal functions, spatial factor analysis, and ARIMA models. To demonstrate, we fit to data from two survey platforms sampling corals, sponges, rockfishes, and flatfishes in the Gulf of Alaska and Aleutian Islands. We then compare eight alternative model structures using different assumptions about habitat drivers and survey detectability. Model selection suggests that towed-camera and bottom trawl gears have spatial variation in detectability but sample the same underlying density of flatfishes and rockfishes, and that rockfishes are positively associated with sponges while flatfishes are negatively associated with corals. We conclude that tinyVAST can be used to test complicated dependencies representing alternative structural assumptions for research and real-world policy evaluation.

2401.10194

Osten Anderson

Osten Anderson and Wanshi Hong and Bin Wang and Nanpeng Yu

Impact of Flexible and Bidirectional Charging in Medium- and Heavy-Duty Trucks on California's Decarbonization Pathway

eess.SY cs.SY

http://creativecommons.org/licenses/by-nc-nd/4.0/

California has committed to ambitious decarbonization targets across multiple sectors, including decarbonizing the electrical grid by 2045. In addition, the medium- and heavy-duty truck fleets are expected to see rapid electrification over the next two decades. Considering these two pathways in tandem is critical for ensuring cost optimality and reliable power system operation. In particular, we examine the potential cost savings of electrical generation infrastructure by enabling flexible charging and bidirectional charging for these trucks. We also examine costs adjacent to enabling these services, such as charger upgrades and battery degradation. We deploy a large mixed-integer decarbonization planning model to quantify the costs associated with the electric generation decarbonization pathway. Example scenarios governing truck driving and charging behaviors are implemented to reveal the sensitivity of temporal driving patterns. Our experiments show that cost savings on the order of multiple billions of dollars are possible by enabling flexible and bidirectional charging in medium- and heavy-duty trucks in California.

2401.10195

Jingyi Wang

Jingyi Wang, Erin Kara, Jeroen Homan, James F. Steiner, Diego Altamirano, Tomaso Belloni, Michiel van der Klis, Adam Ingram, Javier A. Garc\'ia, Guglielmo Mastroserio, Riley Connors, Matteo Lucchini, Thomas Dauser, Joseph Neilsen, Collin Lewin, Ron A. Remillard

Highly-coherent quasi-periodic oscillations in the 'heartbeat' black hole X-ray binary IGR J17091-3624

11 pages, 10 figures, 2 tables, accepted to be published in ApJ

astro-ph.HE

http://creativecommons.org/licenses/by/4.0/

IGR J17091-3624 is a black hole X-ray binary (BHXB), often referred to as the 'twin' of GRS 1915+105 because it is the only other known BHXB that can show exotic 'heartbeat'-like variability that is highly structured and repeated. Here we report on observations of IGR J17091-3624 from its 2022 outburst, where we detect an unusually coherent quasi-periodic oscillation (QPO) when the broadband variability is low (total fractional rms $\lesssim$ 6%) and the spectrum is dominated by the accretion disk. Such spectral and variability behavior is characteristic of the soft state of typical BHXBs (i.e., those that do not show heartbeats), but we also find that this QPO is strongest when there is some exotic heartbeat-like variability (so-called Class V variability). This QPO is detected at frequencies between 5 and 8 Hz and has Q-factors (defined as the QPO frequency divided by the width) $\gtrsim$ 50, making it one of the most highly coherent low-frequency QPO ever seen in a BHXB. The extremely high Q factor makes this QPO distinct from typical low-frequency QPOs that are conventionally classified into Type-A/B/C QPOs. Instead, we find evidence that archival observations of GRS 1915+105 also showed a similarly high-coherence QPO in the same frequency range, suggesting that this unusually coherent and strong QPO may be unique to BHXBs that can exhibit 'heartbeat'-like variability.

2401.10196

Rita Fici

Rita Fici, Luigi Augugliaro and Ernst-Jan Camiel Wit

Functional Conditional Gaussian Graphical Models

19 pages

stat.ME

http://creativecommons.org/licenses/by/4.0/

Functional data has become a commonly encountered data type. In this paper, we contribute to the literature on functional graphical modelling by extending the notion of conditional Gaussian Graphical models and proposing a double-penalized estimator by which to recover the edge-set of the corresponding graph. Penalty parameters play a crucial role in determining the precision matrices for the response variables and the regression matrices. The performance and model selection process in the proposed framework are investigated using information criteria. Moreover, we propose a novel version of the Kullback-Leibler cross-validation designed for conditional joint Gaussian Graphical Models. The evaluation of model performance is done in terms of Kullback-Leibler divergence and graph recovery power.

2401.10197

Martin Houde

Martin Houde and Nicol\'as Quesada

Perfect pulsed inline twin-beam squeezers

14 pages, 3 figures, 3 tables

quant-ph physics.optics

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Perfect inline squeezers are both spectrally pure and have identical input and output temporal modes, allowing one to squeeze an arbitrary input quantum state in the sole input mode on which the device acts, while the quantum states of any other modes are unaffected. We study theoretically how to obtain a perfect pulsed inline squeezer in twin-beam systems by considering three commonly used configurations: unpoled single pass, poled single pass, and poled double pass. By obtaining analytical relations between the input and output temporal modes from the Bloch-Messiah decomposition of the discretized Heisenberg-picture propagator, we find that a double pass structure produces a perfect pulsed inline squeezer when operated in a frequency degenerate, symmetric group-velocity matched type-II configuration.

2401.10198

Yairon Cid-Ruiz

Yairon Cid-Ruiz

Polar multiplicities and integral dependence

math.AC math.AG

http://creativecommons.org/licenses/by/4.0/

We provide new criteria for the integrality and birationality of an extension of graded algebras in terms of the general notion of polar multiplicities of Kleiman and Thorup. As an application, we obtain a new criterion for when a module is a reduction of another in terms of certain mixed Buchsbaum-Rim multiplicities. Furthermore, we prove several technical results regarding polar multiplicities.

2401.10199

Oleg Aristov

O. Yu. Aristov

Envelopes in the class of Banach algebras of polynomial growth and $C^\infty$-functions of a finite number of free variables

math.FA math.OA

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We introduce the notion of envelope of a topological algebra (in particular, an arbitrary associative algebra) with respect to a class of Banach algebra. In the case of the class of real Banach algebras of polynomial growth, i.e., admitting a $C^\infty$-functional calculus for every element, we get a functor that maps the algebra of polynomials in $k$ variables to the algebra of $C^\infty$-functions on $\mathbb{R}^k$. The envelope of a general commutative or non-commutative algebra can be treated as an algebra of $C^\infty$-functions on some commutative or non-commutative space. In particular, we describe the envelopes of the universal enveloping algebra of finite-dimensional Lie algebras, the coordinate algebras of the quantum plane and quantum $SL(2)$ and also look at some commutative examples. A result on algebras of `free $C^\infty$-functions', i.e., the envelopes of free associative algebras of finite rank $k$, is announced for general $k$ and proved for $k\le 2$.

2401.10200

James Bartusek

James Bartusek, Zvika Brakerski, Vinod Vaikuntanathan

Quantum State Obfuscation from Classical Oracles

quant-ph cs.CR

http://creativecommons.org/publicdomain/zero/1.0/

A major unresolved question in quantum cryptography is whether it is possible to obfuscate arbitrary quantum computation. Indeed, there is much yet to understand about the feasibility of quantum obfuscation even in the classical oracle model, where one is given for free the ability to obfuscate any classical circuit. In this work, we develop a new array of techniques that we use to construct a quantum state obfuscator, a powerful notion formalized recently by Coladangelo and Gunn (arXiv:2311.07794) in their pursuit of better software copy-protection schemes. Quantum state obfuscation refers to the task of compiling a quantum program, consisting of a quantum circuit $C$ with a classical description and an auxiliary quantum state $\ket{\psi}$, into a functionally-equivalent obfuscated quantum program that hides as much as possible about $C$ and $\ket{\psi}$. We prove the security of our obfuscator when applied to any pseudo-deterministic quantum program, i.e. one that computes a (nearly) deterministic classical input / classical output functionality. Our security proof is with respect to an efficient classical oracle, which may be heuristically instantiated using quantum-secure indistinguishability obfuscation for classical circuits. Our result improves upon the recent work of Bartusek, Kitagawa, Nishimaki and Yamakawa (STOC 2023) who also showed how to obfuscate pseudo-deterministic quantum circuits in the classical oracle model, but only ones with a completely classical description. Furthermore, our result answers a question of Coladangelo and Gunn, who provide a construction of quantum state indistinguishability obfuscation with respect to a quantum oracle. Indeed, our quantum state obfuscator together with Coladangelo-Gunn gives the first candidate realization of a ``best-possible'' copy-protection scheme for all polynomial-time functionalities.

2401.10201

Joseph Hoisington

Joseph Hoisington

Energy-minimizing mappings of real projective spaces

7 pages, comments welcome

math.DG

http://creativecommons.org/licenses/by/4.0/

We establish upper and lower bounds for the infimum of the energy in homotopy classes of mappings from real projective spaces to Riemannian manifolds. We also show, as a corollary of the work of White in [Wh86], that the infimum of the energy in a homotopy class of mappings of real projective $n$-space is determined by an associated class of mappings of the real projective plane.

2401.10202

Mariia Khelashvili

Mariia Khelashvili, Anton Rudakovskyi and Sabine Hossenfelder

SPARC galaxies prefer Dark Matter over MOND

10 pages, 7 figures

astro-ph.CO astro-ph.GA

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

We currently have two different hypotheses to solve the missing mass problem: dark matter (DM) and modified Newtonian dynamics (MOND). In this work, we use Bayesian inference applied to the Spitzer Photometry and Accurate Rotation Curves (SPARC) galaxies' rotation curves to see which hypothesis fares better. For this, we represent DM by two widely used cusped and cored profiles, Navarro-Frenk-White (NFW) and Burkert. We parameterize MOND by a widely used radial-acceleration relation (RAR). Our results show a preference for the cored DM profile with high Bayes factors in a substantial fraction of galaxies. Interestingly enough, MOND is typically preferred by those galaxies which lack precise rotation curve data. Our study also confirms that the choice of prior has a significant impact on the credible interval of the characteristic MOND acceleration. Overall, our analysis comes out in favor of dark matter.

2401.10203

Laurence Boxer

Laurence Boxer

Remarks on Fixed Point Assertions in Digital Topology, 8

There is a lot of text quoted or paraphrased from other papers. Sources are given for all such text

math.GT

http://creativecommons.org/licenses/by/4.0/

This paper continues a series in which we study deficiencies in previously published works concerning fixed point assertions for digital images.

2401.10204

Maximilian Egger

Maximilian Egger, Rawad Bitar, Antonia Wachter-Zeh, Deniz G\"und\"uz and Nir Weinberger

Maximal-Capacity Discrete Memoryless Channel Identification

cs.IT math.IT stat.ML

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The problem of identifying the channel with the highest capacity among several discrete memoryless channels (DMCs) is considered. The problem is cast as a pure-exploration multi-armed bandit problem, which follows the practical use of training sequences to sense the communication channel statistics. A capacity estimator is proposed and tight confidence bounds on the estimator error are derived. Based on this capacity estimator, a gap-elimination algorithm termed BestChanID is proposed, which is oblivious to the capacity-achieving input distribution and is guaranteed to output the DMC with the largest capacity, with a desired confidence. Furthermore, two additional algorithms NaiveChanSel and MedianChanEl, that output with certain confidence a DMC with capacity close to the maximal, are introduced. Each of those algorithms is beneficial in a different regime and can be used as a subroutine in BestChanID. The sample complexity of all algorithms is analyzed as a function of the desired confidence parameter, the number of channels, and the channels' input and output alphabet sizes. The cost of best channel identification is shown to scale quadratically with the alphabet size, and a fundamental lower bound for the required number of channel senses to identify the best channel with a certain confidence is derived.

2401.10205

Jose Nelson Amaral

Jos\'e Nelson Amaral

Effective Communication of Scientific Results

14 pages manuscript

cs.DL

http://creativecommons.org/licenses/by/4.0/

Communication is essential for the advancement of Science. Technology advances and the proliferation of personal devices have changed the ways in which people communicate in all aspects of life. Scientific communication has also been profoundly affected by such changes, and thus it is important to reflect on effective ways to communicate scientific results to scientists that are flooded with information. This article advocates for receiver-oriented communication in Science, discusses how effective oral presentations should be prepared and delivered, provides advice on the thought process that can lead to scientific papers that communicate effectively, discusses suitable methodology to produce experimental data that is relevant and offers advice on how to present such data in ways that lead to the formulation of correct claims that are supported by the data.

2401.10206

David Sumner Lipham

David Lipham, Jan van Mill, Murat Tuncali, Ed Tymchatyn, Kirsten Valkenburg

Buried points of plane continua

7 pages, 2 figures

math.GN

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sets on the boundary of a complementary component of a continuum in the plane have been of interest since the early 1920's. Curry and Mayer defined the buried points of a plane continuum to be the points in the continuum which were not on the boundary of any complementary component. Motivated by their investigations of Julia sets, they asked what happens if the set of buried points of a plane continuum is totally disconnected and non-empty. Curry, Mayer and Tymchatyn showed that in that case the continuum is Suslinian, i.e. it does not contain an uncountable collection of non-degenerate pairwise disjoint subcontinua. In an answer to a question of Curry et al, van Mill and Tuncali constructed a plane continuum whose buried point set was totally disconnected, non-empty and one-dimensional at each point of a countably infinite set. In this paper we show that the van Mill-Tuncali example was best possible in the sense that whenever the buried set is totally disconnected, then it is one-dimensional at each of at most countably many points. As a corollary we find that the buried set cannot be almost zero-dimensional unless it is zero-dimensional. We also construct locally connected van Mill-Tuncali type examples.

2401.10207

Sudip Mittal

Jesse Ables, Nathaniel Childers, William Anderson, Sudip Mittal, Shahram Rahimi, Ioana Banicescu, Maria Seale

Eclectic Rule Extraction for Explainability of Deep Neural Network based Intrusion Detection Systems

cs.CR cs.AI cs.LG

http://creativecommons.org/licenses/by-nc-nd/4.0/

This paper addresses trust issues created from the ubiquity of black box algorithms and surrogate explainers in Explainable Intrusion Detection Systems (X-IDS). While Explainable Artificial Intelligence (XAI) aims to enhance transparency, black box surrogate explainers, such as Local Interpretable Model-Agnostic Explanation (LIME) and SHapley Additive exPlanation (SHAP), are difficult to trust. The black box nature of these surrogate explainers makes the process behind explanation generation opaque and difficult to understand. To avoid this problem, one can use transparent white box algorithms such as Rule Extraction (RE). There are three types of RE algorithms: pedagogical, decompositional, and eclectic. Pedagogical methods offer fast but untrustworthy white-box explanations, while decompositional RE provides trustworthy explanations with poor scalability. This work explores eclectic rule extraction, which strikes a balance between scalability and trustworthiness. By combining techniques from pedagogical and decompositional approaches, eclectic rule extraction leverages the advantages of both, while mitigating some of their drawbacks. The proposed Hybrid X-IDS architecture features eclectic RE as a white box surrogate explainer for black box Deep Neural Networks (DNN). The presented eclectic RE algorithm extracts human-readable rules from hidden layers, facilitating explainable and trustworthy rulesets. Evaluations on UNSW-NB15 and CIC-IDS-2017 datasets demonstrate the algorithm's ability to generate rulesets with 99.9% accuracy, mimicking DNN outputs. The contributions of this work include the hybrid X-IDS architecture, the eclectic rule extraction algorithm applicable to intrusion detection datasets, and a thorough analysis of performance and explainability, demonstrating the trade-offs involved in rule extraction speed and accuracy.

2401.10208

Jifeng Dai

Changyao Tian, Xizhou Zhu, Yuwen Xiong, Weiyun Wang, Zhe Chen, Wenhai Wang, Yuntao Chen, Lewei Lu, Tong Lu, Jie Zhou, Hongsheng Li, Yu Qiao, Jifeng Dai

MM-Interleaved: Interleaved Image-Text Generative Modeling via Multi-modal Feature Synchronizer

20 pages, 9 figures, 17 tables

cs.CV cs.CL

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Developing generative models for interleaved image-text data has both research and practical value. It requires models to understand the interleaved sequences and subsequently generate images and text. However, existing attempts are limited by the issue that the fixed number of visual tokens cannot efficiently capture image details, which is particularly problematic in the multi-image scenarios. To address this, this paper presents MM-Interleaved, an end-to-end generative model for interleaved image-text data. It introduces a multi-scale and multi-image feature synchronizer module, allowing direct access to fine-grained image features in the previous context during the generation process. MM-Interleaved is end-to-end pre-trained on both paired and interleaved image-text corpora. It is further enhanced through a supervised fine-tuning phase, wherein the model improves its ability to follow complex multi-modal instructions. Experiments demonstrate the versatility of MM-Interleaved in recognizing visual details following multi-modal instructions and generating consistent images following both textual and visual conditions. Code and models are available at \url{https://github.com/OpenGVLab/MM-Interleaved}.

2401.10209

Farnaz Rezaee

Farnaz Rezay Morsagh, Mehdi Siahi, Soudabeh Soleymani

Synchronization and Control of Chaotic Spur Gear System Using Type-II Fuzzy Controller Optimized via Whale Optimization Algorithm

eess.SY cs.SY

http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Interval type-II Fuzzy Inference System (FIS) assumes a crucial role in determining the coefficients of the PID controller, thereby augmenting the controller's flexibility. Controlling chaotic systems presents inherent challenges and difficulties due to their sensitivity to initial conditions and the intricate dynamics that require precise and adaptive control strategies. This paper offers an exhaustive exploration into the coordination and regulation of a chaotic spur gear system, employing a Type-II Fuzzy Controller. The initial control parameters of the PID controller undergo optimization using the Whale Optimization Algorithm (WOA) to increase the overall system performance. The adaptability and strength of the suggested control system are tested in various scenarios, covering diverse reference inputs and uncertainties. The investigation comprehensively assesses the operational efficacy of the formulated controller, contrasting its performance with other methodologies. The outcomes highlight the impressive efficiency of the suggested strategy, confirming its supremacy in attaining synchronization and control within the turbulent spur gear system under demanding circumstances

2401.10210

Anup Shakya

Anup Shakya, Vasile Rus, Deepak Venugopal

Mastery Guided Non-parametric Clustering to Scale-up Strategy Prediction

Proceedings of 37th AAAI Conference on Artificial Intelligence Artificial Intelligence for Education. arXiv admin note: substantial text overlap with arXiv:2308.03892

cs.CY cs.AI cs.LG

http://creativecommons.org/publicdomain/zero/1.0/

Predicting the strategy (sequence of concepts) that a student is likely to use in problem-solving helps Adaptive Instructional Systems (AISs) better adapt themselves to different types of learners based on their learning abilities. This can lead to a more dynamic, engaging, and personalized experience for students. To scale up training a prediction model (such as LSTMs) over large-scale education datasets, we develop a non-parametric approach to cluster symmetric instances in the data. Specifically, we learn a representation based on Node2Vec that encodes symmetries over mastery or skill level since, to solve a problem, it is natural that a student's strategy is likely to involve concepts in which they have gained mastery. Using this representation, we use DP-Means to group symmetric instances through a coarse-to-fine refinement of the clusters. We apply our model to learn strategies for Math learning from large-scale datasets from MATHia, a leading AIS for middle-school math learning. Our results illustrate that our approach can consistently achieve high accuracy using a small sample that is representative of the full dataset. Further, we show that this approach helps us learn strategies with high accuracy for students at different skill levels, i.e., leveraging symmetries improves fairness in the prediction model.