检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Shoutik Mukherjee Peter Jendrichovksy Patrick O. Kanold Behtash Babadi Department of Electrical & Computer Engineering University of Maryland College Park Institute for Systems Research University of Maryland College Park Department of Biomedical Engineering Johns Hopkins University Kavli Neuroscience Discovery Institute Johns Hopkins University

Optogenetic stimulation has opened up a new avenue to probe neuronal circuitry at high spatiotemporal resolutions. A key challenge in optogenetic stimulation is deciding which subset out of thousands of neurons should be stimulated to elicit a desired network activation or affect behavior. In this work, we introduce a reinforcement learning approach to adaptively narrow down the multitude of stimulation possibilities and robustly identify Granger causal networks that underlie neuronal activity. We use realistic simulations with different underlying circuitry to show the effectiveness of reinforcement learning in identifying an optimal policy for selecting stimulation targets.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Investigation of Effective Electrode Structure for Metal Removal from Composite Materials Using Pulsed Discharge

Investigation of Effective Electrode Structure for Metal Rem...

引用

Pulsed Power Conference, International Conference on High-Power Particle Beams and International Symposium on Electromagnetic Launch Technology (EAPPC/BEAMS/EML), Euro-Asian

作者： Takuto Doi Shota Oishi Takashi Sakugawa Kaisei Nishimura Shinichiro Shobako Shinichi Shimasaki Tomohiko Yamashita Department of Computer Science & Electrical Engineering Graduate School of Science & Technology Kumamoto University Kumamoto Japan Department of Electro-Mechanical Systems Engineering National Institute of Technology Kagawa College Takamatsu Japan

ISBN: (数字)9789038661353

ISBN: (纸本)9798350352733

The authors investigate the potential of pulsed power technology in recycling of E-waste. Applying the pulsed discharge can separate composite materials into plastic and metal. In this study, pulsed discharge was applied to indium tin oxide (ITO) coated plastic films. We used two electrode types: rod-to-rod electrode and a pair of flat plate electrodes to investigate the effect of electrode structure on the metal removal area in detail. A series of single pulse discharges were applied at various electrode distances. As a result, it was revealed that metal can be removed over a wide range by the pulsed discharge using a pair of flat plate electrodes. When the gap between the electrodes was 30 mm, the removal area by the flat plate electrodes was approximately 3.4 times that by the rod electrodes. Analysis of the current density also revealed that the metal removal area was greatly affected by the current density.

关键词： Electrodes Particle beams Composite materials Shape Indium tin oxide Metals Plastic films Discharges (electric) Recycling Current density

来源：评论

学校读者我要写书评

暂无评论

Sensor Fault Detection and Isolation in Autonomous Nonlinear systems Using Neural Network-Based Observers

Sensor Fault Detection and Isolation in Autonomous Nonlinear...

引用

European Control Conference (ECC)

作者： John Cao M. Umar B. Niazi Matthieu Barreau Karl Henrik Johansson Division of Decision and Control Systems Digital Futures KTH Royal Institute of Technology Stockholm Sweden Department of Electrical Engineering and Computer Science Laboratory for Information & Decision Systems Massachusetts Institute of Technology Cambridge MA USA

ISBN: (数字)9783907144107

ISBN: (纸本)9798331540920

This paper presents a novel observer-based approach to detect and isolate faulty sensors in nonlinear systems. The proposed sensor fault detection and isolation (s-FDI) method applies to a general class of nonlinear systems. Our focus is on s-FDI for two types of faults: complete failure and sensor degradation. The key aspect of this approach lies in the utilization of a neural network-based Kazantzis-Kravaris/Luenberger (KKL) observer. The neural network is trained to learn the dynamics of the observer, enabling accurate output predictions of the system. Sensor faults are detected by comparing the actual output measurements with the predicted values. If the difference surpasses a theoretical threshold, a sensor fault is detected. To identify and isolate which sensor is faulty, we compare the numerical difference of each sensor measurement with an empirically derived threshold. We derive both theoretical and empirical thresholds for detection and isolation, respectively. Notably, the proposed approach is robust to measurement noise and system uncertainties. Its effectiveness is demonstrated through numerical simulations of sensor faults in a network of Kuramoto oscillators.

关键词： Uncertainty Fault detection Noise Observers Numerical simulation Sensor systems Sensors

来源：评论

学校读者我要写书评

暂无评论

ELA: Exploited Level Augmentation for Offline Learning in Zero-Sum Games

arXiv

引用

arXiv 2024年

作者： Lei, Shiqi Lee, Kanghoon Li, Linjing Park, Jinkyoo Li, Jiachen Institute of Automation Chinese Academy of Sciences Beijing China Department of Industrial & Systems Engineering KAIST Daejeon Korea Republic of Department of Electrical and Computer Engineering University of California RiversideCA United States

Offline learning has become widely used due to its ability to derive effective policies from offline datasets gathered by expert demonstrators without interacting with the environment directly. Recent research has explored various ways to enhance offline learning efficiency by considering the characteristics (e.g., expertise level or multiple demonstrators) of the dataset. However, a different approach is necessary in the context of zero-sum games, where outcomes vary significantly based on the strategy of the opponent. In this study, we introduce a novel approach that uses unsupervised learning techniques to estimate the exploited level of each trajectory from the offline dataset of zero-sum games made by diverse demonstrators. Subsequently, we incorporate the estimated exploited level into the offline learning to maximize the influence of the dominant strategy. Our method enables interpretable exploited level estimation in multiple zero-sum games and effectively identifies dominant strategy data. Also, our exploited level augmented offline learning significantly enhances the original offline learning algorithms including imitation learning and offline reinforcement learning for zero-sum games. Copyright © 2024, 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

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

Distributed Optimization with Gradient Descent and Quantized Communication

引用

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

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

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

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

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

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：