检索结果-内蒙古大学图书馆

DFingerNet: Noise-Adaptive Speech Enhancement for Hearing Aids

学校读者我要写书评

暂无评论

DFingerNet: Noise-Adaptive Speech Enhancement for Hearing Ai...

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Tsangko, Iosif Triantafyllopoulos, Andreas Müller, Michael Schröter, Hendrik Schuller, Björn W. EIHW - Embedded Intelligence for Health Care and Wellbeing University of Augsburg Germany Technical University of Munich Germany MCML - Munich Center for Machine Learning Munich Germany WS Audiology Research and Development Erlangen Germany GLAM - Group on Language Audio & Music Imperial College London United Kingdom MDSI - Munich Data Science Institute Munich Germany

ISBN: (纸本)9798350368741

The DeepFilterNet (DFN) architecture was recently proposed as a deep learning model suited for hearing aid devices. Despite its competitive performance on numerous benchmarks, it still follows a 'one-size-fits-all' approach, which aims to train a single, monolithic architecture that generalises across different noises and environments. However, its limited size and computation budget can hamper its generalisability. Recent work has shown that in-context adaptation can improve performance by conditioning the denoising process on additional information extracted from background recordings to mitigate this. These recordings can be offloaded outside the hearing aid, thus improving performance while adding minimal computational overhead. We introduce these principles to the DFN model, thus proposing the DFingerNet (DFiN) model, which shows superior performance on various benchmarks inspired by the DNS Challenge. © 2025 IEEE.

关键词： Context Adaptation Deep learning Hearing Aids Speech Denoising Speech Enhancement

RESPONSIBLE AI AGENTS

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Desai, Deven R. Riedl, Mark O. Georgia Institute of Technology J.D. Yale Law School Yale Law School Information Society Project former Academic Research Counsel Google Inc. Georgia Tech Machine Learning Center United States Georgia Tech School of Interactive Computing Georgia Institute of Technology Georgia Tech Machine Learning Center United States

Thanks to advances in large language models, a new type of software agent, the artificial intelligence (AI) agent, has entered the marketplace. Companies such as OpenAI, Google, Microsoft, and Salesforce promise their AI Agents will go from generating passive text to executing tasks. Instead of a travel itinerary, an AI Agent would book all aspects of your trip. Instead of generating text or images for social media post, an AI Agent would post the content across a host of social media outlets. The potential power of AI Agents has fueled legal scholars’ fears that AI Agents will enable rogue commerce, human manipulation, rampant defamation, and intellectual property harms. These scholars are calling for regulation before AI Agents cause havoc. This Article addresses the concerns around AI Agents head on. It shows that core aspects of how one piece of software interacts with another creates ways to discipline AI Agents so that rogue, undesired actions are unlikely, perhaps more so than rules designed to govern human agents. It also develops a way to leverage the computer-science approach to value-alignment to improve a user’s ability to take action to prevent or correct AI Agent operations. That approach offers and added benefit of helping AI Agents align with norms around user-AI Agent interactions. These practices will enable desired economic outcomes and mitigate perceived risks. The Article also argues that no matter how much AI Agents seem like human agents, they need not, and should not, be given legal personhood status. In short, humans are responsible for AI Agents’ actions, and this Article provides a guide for how humans can build and maintain responsible AI Agents. © 2025, CC BY.

关键词： Software agents

Point Cloud Neural Operator for Parametric PDEs on Complex and Variable Geometries

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zeng, Chenyu Zhang, Yanshu Zhou, Jiayi Wang, Yuhan Wang, Zilin Liu, Yuhao Wu, Lei Huang, Daniel Zhengyu School of Mathematical Sciences Peking University Beijing China Department of Mathematical Sciences Tsinghua University Beijing China Center for Machine Learning Research Peking University Beijing China Beijing International Center for Mathematical Research Peking University Beijing China

Surrogate models are critical for accelerating computationally expensive simulations in science and engineering, particularly for solving parametric partial differential equations (PDEs). Key challenges in developing practical surrogate models include managing high-dimensional inputs and outputs and handling geometrically complex and variable domains, which are often represented as point clouds. In this work, we systematically investigate the formulation of neural operators on point clouds and introduce the Point Cloud Neural Operator (PCNO), a neural network-based surrogate model designed to efficiently approximate solution maps of parametric PDEs on complex and variable geometries. We evaluate the performance of PCNO on a range of pedagogical PDE problems, focusing on aspects such as boundary layers, adaptively meshed point clouds, and variable domains with topological variations. Its practicality is further demonstrated through three-dimensional applications, such as predicting pressure loads on various types of vehicles and simulating the inflation process of intricate parachute structures. © 2025, CC BY.

关键词： Boundary layers

Bayesian Optimization for Building Social-Influence-Free Consensus

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Adachi, Masaki Chau, Siu Lun Xu, Wenjie Singh, Anurag Osborne, Michael A. Muandet, Krikamol Machine Learning Research Group University of Oxford United Kingdom Toyota Motor Corporation Japan CISPA Helmholtz Center for Information Security Germany Automatic Control Laboratory EPFL Switzerland

We introduce Social Bayesian Optimization (SBO), a vote-efficient algorithm for consensus-building in collective decision-making. In contrast to single-agent scenarios, collective decision-making encompasses group dynamics that may distort agents’ preference feedback, thereby impeding their capacity to achieve a social-influence-free consensus—the most preferable decision based on the aggregated agent utilities. We demonstrate that under mild rationality assumptions, reaching social-influence-free consensus using noisy feedback alone is impossible. To address this, SBO employs a dual voting system: cheap but noisy public votes (e.g., show of hands in a meeting), and more accurate, though expensive, private votes (e.g., one-to-one interview). We model social influence using an unknown social graph and leverage the dual voting system to efficiently learn this graph. Our theoretical findings show that social graph estimation converges faster than the black-box estimation of agents’ utilities, allowing us to reduce reliance on costly private votes early in the process. This enables efficient consensus-building primarily through noisy public votes, which are debiased based on the estimated social graph to infer social-influence-free feedback. We validate the efficacy of SBO across multiple real-world applications, including thermal comfort, team building, travel negotiation, and energy trading *** Codes 62C10, 62F15 © 2025, CC BY.

关键词： Voting machines

Generating a Question Answering Dataset About Geographic Changes in a Knowledge Graph 24th

学校读者我要写书评

暂无评论

Generating a Question Answering Dataset About Geographic Ch...

24th International Conference on Knowledge Engineering and Knowledge Management, EKAW 2024

作者： Mitsios, Michalis Punjani, Dharmen Abdollahi, Sara Gottschalk, Simon Tsalapati, Eleni Demidova, Elena Koubarakis, Manolis AI Team Department of Informatics and Telecommunications National and Kapodistrian University of Athens Athens Greece Hubert Curien Laboratory Université Jean Monnet Saint Etienne France L3S Research Center Leibniz Universität Hannover Hanover Germany Athens Technology Center Chalandri Greece University of Bonn Bonn Germany Lamarr Institute for Machine Learning and Artificial Intelligence Bonn Germany Archimedes/Athena RC Marousi Greece

ISBN: (纸本)9783031777912

Most studies on semantic question answering (QA) are predominantly focused on encyclopedic knowledge graphs like DBpedia and Wikidata. These studies cover, if at all, the spatial and temporal characteristics of geospatial entities in isolation, not addressing them simultaneously. In this paper, we introduce a pipeline for creating question answering datasets for evaluating the reasoning capabilities of QA models in the context of geographic changes over time. This pipeline generates questions, GeoSPARQL queries, and corresponding answers by leveraging subgraph and query template extraction techniques. We exemplify this pipeline with the creation of the GeoChangesQA dataset with questions over a knowledge graph of US counties and states and their changes from 1629 to 2000. By evaluating GeoChangesQA using a Transformer-based model, we demonstrate that historical geospatial questions pose a substantial challenge for semantic question answering. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Question answering

E-3SFC: Communication-Efficient Federated learning with Double-way Features Synthesizing

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zhou, Yuhao Tian, Yuxin Shi, Mingjia Li, Yuanxi Sun, Yanan Ye, Qing Lv, Jiancheng The College of Computer Science Sichuan University Chengdu610065 China Engineering Research Center of Machine Learning and Industry Intelligence Ministry of Education Chengdu610065 China The University of Illinois at Urbana-Champaign United States

The exponential growth in model sizes has significantly increased the communication burden in Federated learning (FL). Existing methods to alleviate this burden by transmitting compressed gradients often face high compression errors, which slow down the model’s convergence. To simultaneously achieve high compression effectiveness and lower compression errors, we study the gradient compression problem from a novel perspective. Specifically, we propose a systematical algorithm termed Extended Single-Step Synthetic Features Compressing (E-3SFC), which consists of three sub-components, i.e., the Single-Step Synthetic Features Compressor (3SFC), a double-way compression algorithm, and a communication budget scheduler. First, we regard the process of gradient computation of a model as decompressing gradients from corresponding inputs, while the inverse process is considered as compressing the gradients. Based on this, we introduce a novel gradient compression method termed 3SFC, which utilizes the model itself as a decompressor, leveraging training priors such as model weights and objective functions. 3SFC compresses raw gradients into tiny synthetic features in a single-step simulation, incorporating error feedback to minimize overall compression errors. To further reduce communication overhead, 3SFC is extended to E-3SFC, allowing double-way compression and dynamic communication budget scheduling. Our theoretical analysis under both strongly convex and non-convex conditions demonstrates that 3SFC achieves linear and sub-linear convergence rates with aggregation noise. Extensive experiments across six datasets and six models reveal that 3SFC outperforms state-of-the-art methods by up to 13.4% while reducing communication costs by 111.6 times. These findings suggest that 3SFC can significantly enhance communication efficiency in FL without compromising model performance. © 2025, CC0.

关键词： Federated learning

Wheel-GINS: A GNSS/INS Integrated Navigation System with a Wheel-mounted IMU

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wu, Yibin Kuang, Jian Niu, Xiaoji Stachniss, Cyrill Klingbeil, Lasse Kuhlmann, Heiner Center for Robotics Institute of Geodesy and Geoinformation University of Bonn Bonn Germany Department of Engineering Science University of Oxford United Kingdom Lamarr Institute for Machine Learning and Artificial Intelligence Germany GNSS Research Center Wuhan University Wuhan China

A long-term accurate and robust localization system is essential for mobile robots to operate efficiently outdoors. Recent studies have shown the significant advantages of the wheelmounted inertial measurement unit (Wheel-IMU)-based dead reckoning system. However, it still drifts over extended periods because of the absence of external correction signals. To achieve the goal of long-term accurate localization, we propose Wheel- GINS, a Global Navigation Satellite System (GNSS)/inertial navigation system (INS) integrated navigation system using a Wheel-IMU. Wheel-GINS fuses the GNSS position measurement with the Wheel-IMU via an extended Kalman filter to limit the long-term error drift and provide continuous state estimation when the GNSS signal is blocked. Considering the specificities of the GNSS/Wheel-IMU integration, we conduct detailed modeling and online estimation of the Wheel-IMU installation parameters, including the Wheel-IMU leverarm and mounting angle and the wheel radius error. Experimental results have shown that Wheel- GINS outperforms the traditional GNSS/Odometer/INS integrated navigation system during GNSS outages. At the same time, Wheel-GINS can effectively estimate the Wheel-IMU installation parameters online and, consequently, improve the localization accuracy and practicality of the system. The source code of our implementation is publicly available (https://***/i2Nav- WHU/Wheel-GINS). Copyright © 2025, The Authors. All rights reserved.

关键词： Extended Kalman filters

STABLE DERIVATIVE FREE GAUSSIAN MIXTURE VARIATIONAL INFERENCE FOR BAYESIAN INVERSE PROBLEMS

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Che, Baojun Chen, Yifan Huan, Zhenghao Huang, Daniel Zhengyu Wang, Weijie School of Mathematical Sciences Nankai University Tianjin China Courant Institute New York University NY United States School of Mathematical Sciences Peking University Beijing China Beijing International Center for Mathematical Research Center for Machine Learning Research Peking University Beijing China

This paper is concerned with the approximation of probability distributions known up to normalization constants, with a focus on Bayesian inference for large-scale inverse problems in scientific computing. In this context, key challenges include costly repeated evaluations of forward models, multimodality, and inaccessible gradients for the forward model. To address them, we develop a variational inference framework that combines Fisher-Rao natural gradient with specialized quadrature rules to enable derivative free updates of Gaussian mixture variational families. The resulting method, termed Derivative Free Gaussian Mixture Variational Inference (DF-GMVI), guarantees covariance positivity and affine invariance, offering a stable and efficient framework for approximating complex posterior distributions. The effectiveness of DF-GMVI is demonstrated through numerical experiments on challenging scenarios, including distributions with multiple modes, infinitely many modes, and curved modes in spaces with up to hundreds of dimensions. The method's practicality is further demonstrated in a large-scale application, where it successfully recovers the initial conditions of the Navier-Stokes equations from solution data at positive times. Copyright © 2025, The Authors. All rights reserved.

关键词： Navier Stokes equations

Online Rack Placement in Large-Scale Data centers

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Baxi, Saumil Cummings, Kayla Jacquillat, Alexandre Lo, Sean McDonald, Rob Mellou, Konstantina Menache, Ishai Molinaro, Marco Cloud Operations and Innovation Microsoft United States Cloud Supply Chain Sustainability Engineering Microsoft United States Sloan School of Management Operations Research Center Massachusetts Institute of Technology United States Machine Learning and Optimization Microsoft Research United States

This paper optimizes the configuration of large-scale data centers toward cost-effective, reliable and sustainable cloud supply chains. We formulate an integer optimization model that optimizes the placement of racks of servers within a data center to maximize demand coverage, adhere to space, power and cooling restrictions, and pace resource utilization for future demand. We also define a tractable single-sample online approximation (SSOA) approach to multi-stage stochastic optimization, which approximates unknown parameters with a single realization and re-optimizes decisions dynamically. Theoretical results provide strong performance guarantees of SSOA in the canonical online generalized assignment and online bin packing settings. Computational results using real-world data show that our optimization approach can enhance utilization and reduce power stranding in data centers. Following iterative improvements in collaboration with data center managers, our algorithm has been packaged into a software solution deployed in Microsoft's data centers worldwide. Deployment data indicate a significant increase in adoption, leading to improved power utilization, multi-million-dollar annual cost savings, and concomitant savings in greenhouse gas emissions. Ultimately, this paper constitutes one of the first large-scale deployments of a decision-making tool in data centers, contributing an interactive decision-making process at the human-machine *** Codes 90C11 Copyright © 2025, The Authors. All rights reserved.

关键词： Chains