检索结果-内蒙古大学图书馆

Distributed Optimization with Gradient Descent and Quantized Communication

IFAC-PapersOnLine 2023年第2期56卷 5900-5906页

作者： Apostolos I. Rikos Wei Jiang Themistoklis Charalambous Karl H. Johansson Division of Decision and Control Systems KTH Royal Institute of Technology SE-100 44 Stockholm Sweden Digital Futures SE-100 44 Stockholm Sweden Department of Electrical Engineering and Automation School of Electrical Engineering Aalto University Espoo Finland Department of Electrical and Computer Engineering School of Engineering University of Cyprus Nicosia Cyprus

In this paper, we consider the unconstrained distributed optimization problem, in which the exchange of information in the network is captured by a directed graph topology, thus, nodes can only communicate with their neighbors. Additionally, in our problem, the communication channels among the nodes have limited bandwidth. In order to alleviate this limitation, quantized messages should be exchanged among the nodes. For solving this distributed optimization problem, we combine a gradient descent method with a distributed quantized consensus algorithm (which requires the nodes to exchange quantized messages and converges in a finite number of steps). Specifically, at every optimization step, each node (i) performs a gradient descent step (i.e., subtracts the scaled gradient from its current estimate), and (ii) performs a finite-time calculation of the quantized average of every node's estimate in the network. As a consequence, this algorithm approximately mimics the centralized gradient descent algorithm. We show that our algorithm asymptotically converges to a neighborhood of the optimal solution with linear convergence rate. The performance of the proposed algorithm is demonstrated via simple illustrative examples.

关键词： Distributed optimization quantized communication directed graphs finite-time consensus

来源：评论

学校读者我要写书评

暂无评论

Communication-efficient and Decentralized Federated Minimax Optimization

Communication-efficient and Decentralized Federated Minimax ...

引用

Annual Allerton Conference on Communication, Control, and Computing

作者： Sajjad Ghiasvand Amirhossein Reisizadeh Mahnoosh Alizadeh Ramtin Pedarsani Department of Electrical and Computer Engineering University of California Santa Barbara CA USA Laboratory for Information and Decision Systems Massachusetts Institute of Technology Cambridge MA USA

ISBN: (数字)9798331541033

ISBN: (纸本)9798331541040

As distributed learning applications like Federated Learning, the Internet of Things (IoT), and Edge Computing expand, addressing their limitations becomes crucial. We approach decentralized learning across a network of communicating clients or nodes, focusing on two primary challenges: data heterogeneity and adversarial robustness. To address these, we introduce a decentralized minimax optimization method incorporating two key components: local updates and gradient tracking. Minimax optimization serves as a fundamental tool for adversarial training, ensuring robustness. Local updates are vital in Federated Learning (FL) to alleviate the communication bottleneck, while gradient tracking is necessary to demonstrate convergence amid data heterogeneity. Our analysis of the proposed algorithm, Dec-Fed Track, in nonconvex-strongly-concave minimax optimization demonstrates its convergence to a stationary point. Additionally, numerical experiments support our theoretical results.

关键词： Training computer aided instruction Federated learning Distance learning Optimization methods Focusing Robustness Internet of Things Convergence Edge computing

来源：评论

学校读者我要写书评

暂无评论

An Actuator with Magnetic Restoration, Part I: Electromechanical Model and Identification

arXiv

引用

arXiv 2023年

作者： Mohammadi, Sajjad Benner, William R. Kirtley, James L. Lang, Jeffrey H. The Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology CambridgeMA02139 United States The Pangolin Laser Systems

electromechanical models are crucial in the design and control of motors and actuators. Modeling, identification, drive, and current control loop of a limited-rotation actuator with magnetic restoration is presented. New nonlinear and linearized electromechanical models are developed for the design of the drive as well as small and large signal controls of the actuator. To attain a higher accuracy and an efficient design, and the eddy-currents in the laminations and magnet are modeled. This involves analytically solving 1-D and 2-D diffusion equations, leading to the derivation of a lumped-element circuit for system-level analyses, such as control system design. Additionally, the study analyzes and incorporates the impact of pre-sliding friction. The actuator is prototyped, and the paper delves into the identification of the model, presenting a procedure for parameter extraction. A close agreement is observed between the results obtained from the model, finite element analysis, and experimental results. The superiority of the proposed model over previous approaches is highlighted. Part II of the paper is dedicated to the drive circuit, the current control, as well as linear and nonlinear position control system designs. Copyright © 2023, The Authors. All rights reserved.

关键词： Finite element method

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning Based Dynamic Power Control for UAV Mobility Management

arXiv

引用

arXiv 2023年

作者： Meer, Irshad A. Besser, Karl-Ludwig Ozger, Mustafa Poor, H. Vincent Cavdar, Cicek Department of Electrical and Computer Engineering Princeton University United States Division of Communication Systems KTH Royal Institute of Technology Sweden

—Modern communication systems need to fulfill multiple and often conflicting objectives at the same time. In particular, new applications require high reliability while operating at low transmit powers. Moreover, reliability constraints may vary over time depending on the current state of the system. One solution to address this problem is to use joint transmissions from a number of base stations (BSs) to meet the reliability requirements. However, this approach is inefficient when considering the overall total transmit power. In this work, we propose a reinforcement learning-based power allocation scheme for an unmanned aerial vehicle (UAV) communication system with varying communication reliability requirements. In particular, the proposed scheme aims to minimize the total transmit power of all BSs while achieving an outage probability that is less than a tolerated threshold. This threshold varies over time, e.g., when the UAV enters a critical zone with high-reliability requirements. Our results show that the proposed learning scheme uses dynamic power allocation to meet varying reliability requirements, thus effectively conserving energy. Copyright © 2023, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Semantic Template-based Convolutional Neural Network for Text Classification

引用

ACM Transactions on Asian and Low-Resource Language Information Processing 2023年第11期22卷 1-21页

作者： Chang, Yung-Chun Ng, Siu Hin Chen, Jung-Peng Liang, Yu-Chi Hsu, Wen-Lian Graduate Institute of Data Science Taipei Medical University Taipei Taiwan Clinical Big Data Research Center Taipei Medical University Hospital Taipei Taiwan Snhcc Tigp Academia Sinica National Cheng Chi University Taipei Taiwan Institute of Information Systems and Applications National Tsing Hua University Hsinchu Taiwan Department of Computer Science and Information Engineering College of Information and Electrical Engineering Asia University Taichung Taiwan Pervasive Ai Research Labs Ministry of Science and Technology Hsinchu City Taiwan

We propose a semantic template-based distributed representation for the convolutional neural network called Semantic Template-based Convolutional Neural Network (STCNN) for text categorization that imitates the perceptual behavior of human comprehension. STCNN is a highly automatic approach that learns semantic templates that characterize a domain from raw text and recognizes categories of documents using a semantic-infused convolutional neural network that allows a template to be partially matched through a statistical scoring system. Our experiment results show that STCNN effectively classifies documents in about 140,000 Chinese news articles into predefined categories by capturing the most prominent and expressive patterns and achieves the best performance among all compared methods for Chinese topic classification. Finally, the same knowledge can be directly used to perform a semantic analysis task. © 2023 held by the owner/author(s). Publication rights licensed to ACM.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Fast yet predictable braking manoeuvers for real-time robot control

引用

IFAC-PapersOnLine 2023年第2期56卷 9984-9991页

作者： Mazin Hamad Jesus Gutierrez-Moreno Hugo T.M. Kussaba Nico Mansfeld Saeed Abdolshah Abdalla Swikir Wolfram Burgard Sami Haddadin Chair of Robotics and Systems Intelligence Munich Institute of Robotics and Machine Intelligence Technical University of Munich Germany Franka Emika GmbH Munich Germany Department of Electrical and Electronic Engineering Omar Al-Mukhtar University Libya Centre for Tactile Internet with Human-in-the-Loop (CeTI) Germany Department of Engineering University of Technology Nuremberg Germany

This paper proposes a framework for generating fast, smooth and predictable braking manoeuvers for a controlled robot. The proposed framework integrates two approaches to obtain feasible modal limits for designing braking trajectories. The first approach is realtime capable but conservative considering the usage of the available feasible actuator control region, resulting in longer braking times. In contrast, the second approach maximizes the used braking control inputs at the cost of requiring more time to evaluate larger, feasible modal limits via optimization. Both approaches allow for predicting the robot's stopping trajectory online. In addition, we also formulated and solved a constrained, nonlinear final-time minimization problem to find optimal torque inputs. The optimal solutions were used as a benchmark to evaluate the performance of the proposed predictable braking framework. A comparative study was compiled in simulation versus a classical optimal controller on a 7-DoF robot arm with only three moving joints. The results verified the effectiveness of our proposed framework and its integrated approaches in achieving fast robot braking manoeuvers with accurate online predictions of the stopping trajectories and distances under various braking settings.

关键词： Controlled stop optimal control braking manoeuvers stopping trajectory prediction

来源：评论

学校读者我要写书评

暂无评论

A Planar Omnidirectional Wireless Power Transfer Platform for Portable Devices

A Planar Omnidirectional Wireless Power Transfer Platform fo...

引用

Annual IEEE Conference on Applied Power Electronics Conference and Exposition (APEC)

作者： Xipei Yu Junjie Feng Qiang Li Bradley Department of Electrical and Computer Engineering Center for Power Electronics Systems Virginia Polytechnic Institute and State University Blacksburg USA

Recently, omnidirectional wireless power transfer (WPT) has gained extensive attention globally. However, it is difficult to integrate such system with common home furniture for its three-dimensional (3-D) structure, which degrades its practicality. Targeting at this challenge, this article proposes a novel planar omnidirectional WPT system, enabling 3-D omnidi-rectional WPT for consumer electronics, such as mobile phones and earphones. It consists of three sets of planar coils, which can generate magnetic field in three dimensions, enabling the receiver to be charged in all positions and orientations. Besides, the decoupling performance of the proposed coil structure is verified. Afterwards, a LCCL-LC resonant converter is utilized to achieve the constant output voltage gain at resonant frequency. Finally, in the experiment, three sets of coils are integrated as PCB windings and the magnetic field is measured to verify the omnidirectional magnetic field distribution under 6.78MHz. When charging a 5-W receiver for portable devices, the output voltage will be between 16-24V while the overall efficiency is 70%-72.5% with $x$ and $y$ misalignment. Considering the angular misalignment, the efficiency remains between 28%-53%.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Speech Separation for Low-Resource Languages

Speech Separation for Low-Resource Languages

引用

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

作者： Borsdorf, Marvin Pan, Zexu Himmelmann, Pascal Li, Haizhou Schultz, Tanja Machine Listening Lab University of Bremen Bremen Germany Alibaba Group Singapore Department of Mathematics and Computer Science FernUniversität in Hagen Hagen Germany School of Data Science Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen China Department of Electrical and Computer Engineering National University of Singapore Singapore Cognitive Systems Lab University of Bremen Bremen Germany

ISBN: (纸本)9798350368741

Speech separation aims to equip machines with the human ability of selective listening, i.e. to focus attention on specific information in spoken communication. Studies have shown that the language spoken in a cocktail party scenario matters. While the development of speech separation models can leverage extensive databases, for the majority of languages only very limited data is available. This work presents the very first study on speech separation for low-resource languages. We choose blind source separation as the task to be studied and analyze three strategies to overcome the data scarcity of two low-resource languages from the GlobalPhoneMS2 database. We show that data from other languages can be used to develop models that work for low-resource languages. Finetuning additionally boosts the performance, and training on multiple languages increases both performance and robustness. We show that dynamic mixing in the development helps to find a trade-off between performance and development time. © 2025 IEEE.

关键词： cocktail party problem GlobalPhone low-resource language multilingual speech separation

来源：评论

学校读者我要写书评

暂无评论

CLEARING SECTIONS OF LATTICE LIABILITY NETWORKS

arXiv

引用

arXiv 2025年

作者： Ghrist, Robert Gould, Julian Lopez, Miguel Riess, Hans Department of Mathematics and Electrical & Systems Engineering University of Pennsylvania Philadelphia PA Department of Mathematics University of Pennsylvania Philadelphia PA Applied Mathematics & Computational Science University of Pennsylvania Philadelphia PA School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta GA

Modern financial networks involve complex obligations that transcend simple monetary debts: multiple currencies, prioritized claims, supply chain dependencies, and more. We present a mathematical framework that unifies and extends these scenarios by recasting the classical Eisenberg-Noe model of financial clearing in terms of lattice liability networks. Each node in the network carries a complete lattice of possible states, while edges encode nominal liabilities. Our framework generalizes the scalar-valued clearing vectors of the classical model to lattice-valued clearing sections, preserving the elegant fixed-point structure while dramatically expanding its descriptive power. Our main theorem establishes that such networks possess clearing sections that themselves form a complete lattice under the product order. This structure theorem enables tractable analysis of equilibria in diverse domains, including multi-currency financial systems, decentralized finance with automated market makers, supply chains with resource transformation, and permission networks with complex authorization structures. We further extend our framework to chain-complete lattices for term structure models and multivalued mappings for complex negotiation systems. Our results demonstrate how lattice theory provides a natural language for understanding complex network dynamics across multiple domains, creating a unified mathematical foundation for analyzing systemic risk, resource allocation, and network *** Codes 91G40, 06B23, 91D30, 68Q85 Copyright © 2025, The Authors. All rights reserved.

关键词： Lattice theory

来源：评论

学校读者我要写书评

暂无评论

Self-excitation of Terahertz Plasmons in Graphene FETs Enabled by Transit-time Negative Dynamic Conductance

Self-excitation of Terahertz Plasmons in Graphene FETs Enabl...

引用

Advancement in Technology (ICONAT), International Conference for

作者： Maxim Ryzhii Victor Ryzhii Taiichi Otsuji Vladimir Mitin Michael S. Shur Department of Computer Science and Engineering University of Aizu Aizu-Wakamatsu Japan Research Institute of Electrical Communication Tohoku University Sendai Japan Department of Electrical Engineering University at Buffalo SUNY Buffalo New York USA Department of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute Troy New York USA

The interband Zener-Klein interband tunneling in the field-effect transistors (FETs) with the graphene channels consisting of the reverse-biased p-i-n junctions enables the negative dynamic conductance associated with the electrons and hole propagation delays. This can lead to the self-excitation of plasmons in the gated p- and n-regions. For the submicrometer gate lengths, the frequency of these plasmons is in the terahertz (THz) range. The plasmon self-excitation in this regime can be used for the generation of THz radiation.

关键词： Ring oscillators Terahertz wave imaging PIN photodiodes Graphene Field effect transistors Logic gates Tunneling

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：