检索结果-内蒙古大学图书馆

Asynchronous Stochastic approximations With Asymptotically Biased Errors and Deep Multiagent Learning

IEEE TRANSACTIONS ON AUTOMATIC CONTROL 2021年第9期66卷 3969-3983页

作者： Ramaswamy, Arunselvan Bhatnagar, Shalabh Quevedo, Daniel E. Paderborn Univ Heinz Nixdorf Inst D-33098 Paderborn Germany Paderborn Univ Dept Comp Sci D-33098 Paderborn Germany Indian Inst Sci Dept Comp Sci & Automat Bangalore 560012 Karnataka India Indian Inst Sci Robert Bosch Ctr Cyber Phys Syst Bangalore 560012 Karnataka India Queensland Univ Technol Sch Elect Engn & Robot Brisbane Qld 4001 Australia

Asynchronous stochastic approximations (SAs) are an important class of model-free algorithms, tools, and techniques that are popular in multiagent and distributed control scenarios. To counter Bellman's curse of dimensionality, such algorithms are coupled with function approximations. Although the learning/control problem becomes more tractable, function approximations affect stability and convergence. In this article, we present verifiable sufficient conditions for stability and convergence of asynchronous SAs with biased approximation errors. The theory developed herein is used to analyze policy gradient methods and noisy value iteration schemes. Specifically, we analyze the asynchronous approximate counterparts of the policy gradient (A2PG) and value iteration (A2VI) schemes. It is shown that the stability of these algorithms is unaffected by biased approximation errors, provided that they are asymptotically bounded. With respect to convergence (of A2VI and A2PG), a relationship between the limiting set and the approximation errors is established. Finally, experimental results that support the theory are presented.

关键词： approximation algorithms approximation error Stability analysis Function approximation Stochastic processes Asymptotic stability Convergence Almost sure boundedness (stability) asymptotically biased approximation errors asynchronous stochastic approximations deep function approximations deep reinforcement learning distributed control multi agent learning networked control systems neuro-dynamic programming

来源：评论

学校读者我要写书评

暂无评论

Multi-Tag UWB-based Indoor Positioning System for Objects Tracking

Multi-Tag UWB-based Indoor Positioning System for Objects Tr...

引用

International Symposium on Intelligent Systems and Informatics (SISY)

作者： Mohammed Faeik Ruzaij Al-Okby Thomas Roddelkopf Simon-Johannes Burgdorf Kerstin Thurow Center for Life Science Automation (celisca) University of Rostock Rostock Germany Technical Institute of Babylon Al-Furat Al-Awsat Technical University (ATU) Kufa Iraq Institute of Automation University of Rostock Rostock Germany

ISBN: (数字)9798350385601

ISBN: (纸本)9798350385618

The work environment in automated laboratories requires indoor positioning systems to track the motion of samples, mobile robots, and workers while implementing their daily tasks. The need for position detection systems lies in achieving maximum levels of early response in case of any emergencies, such as collision accidents or leakage of chemical or biological samples while being transported by robots. The process of tracking objects and determining their precise location is a complex task of great importance in the field of automation. The recent developments in communications and electronics technologies have made this task possible. In this paper, the testing and evaluation of a low-cost indoor positioning system for tracking the movement of mobile objects is presented. The system consists of several commercial ultra-wideband modules that can be used as tags or anchors based on the hosted processor programming. It can be used to track any moving object, such as labware, robots, humans, and other objects by connecting a tracking unit (tag) to the moving object. Practical results strongly suggest that the system is capable of tracking the required targets and displaying their live data with high accuracy and an error rate of less than 50 cm in line-of-sight conditions.

关键词： Target tracking Tracking Error analysis System performance Programming approximation algorithms IP networks Indoor positioning systems Wideband Ultra wideband technology

来源：评论

学校读者我要写书评

暂无评论

A Novel Real-time Algorithm Based on Phase-Locked Data Alignment for Continuously Controlled SSVEP-BCI

A Novel Real-time Algorithm Based on Phase-Locked Data Align...

引用

Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

作者： Hanzhe Jiang Xiaolin Xiao Jie Mei Minpeng Xu Kun Wang Dong Ming Academy of Medical Engineering and Translational Medicine Tianjin University Tianjin People’s Republic of China Haihe Laboratory of Brain-Computer Interaction and Human-Machine Integration Tianjin People’s Republic of China

ISBN: (数字)9798350371499

ISBN: (纸本)9798350371505

Objective: Brain-computer interfaces (BCIs) based on steady-state visual evoked potential (SSVEP) show good performance. However, algorithms always decode segments of electroencephalogram (EEG) and can only satisfy discrete output instructions, which limit the real-time continuous control of the BCI system. This article proposes a novel algorithm for SSVEP-BCI that can translate continuous EEG into control commands, achieving real-time monitoring of user intentions. Methods: A phase synchronicity maximum strategy has been employed in this algorithm, which could capture a fixed-duration SSVEP epoch near any given moment, ensuring each trial is aligned with the phase of the potential corresponding template. Then, the algorithm utilized an update strategy of a small-step sliding window to recognize and output commands in approximately real-time. Results: We constructed an SSVEP-BCI system with continuous stimulation and recruited nine subjects. The results showed that the algorithm proposed in this study efficiently decoded continuously evoked SSVEP signals. The BCI’s online average accuracy and ITR were 92.03% and 143.38 bits/min, respectively. Significance: The proposed algorithm can decode SSVEP at any time theoretically, which improves command output density as well as maintains high recognition accuracy. This study provides novel methods for real-time control of external devices using SSVEP-BCIs and helps to develop BCIs that are more compatible with human control habits.

关键词： Visualization Time-frequency analysis Accuracy approximation algorithms Real-time systems Electroencephalography Decoding Windows Steady-state Engineering in medicine and biology

来源：评论

学校读者我要写书评

暂无评论

Secure and Energy-Efficient Mobile Edge Computing with UAV-Mounted- RIS Assistance

Secure and Energy-Efficient Mobile Edge Computing with UAV-M...

引用

IEEE Conference on Vehicular Technology (VTC)

作者： Ali A. Nasir Department of Electrical Engineering Center for Communication Systems Sensing at King Fahd University of Petroleum and Minerals (KFUPM) Dhahran Saudi Arabia

ISBN: (数字)9798350387414

ISBN: (纸本)9798350387421

This paper studies secure and energy-efficient mobile edge computing with UAV-mounted RIS assistance, in the presence of multiple users (UEs) and an eavesdropper (EV). With an objective of maximizing the minimum energy efficiency (EE) among all UEs in the company of an EV, we jointly optimize parameters including user scheduling, phase-shifts of the programmable reflecting elements, UAV trajectory design, UEs' power allocation, and their computation task and CPU frequency allocation. To ensure secure communication, we adopt power-splitting based artificial noise transmission. We solve this challenging problem by employing successive convex approximation and block coordinate descent techniques. We showcase the superiority of our proposed algorithm over existing studies through simulation results. Particularly, our proposed algorithm manages to achieve the EE very close to the benchmark performance, which assumes the absence of an EV.

关键词： Multi-access edge computing Processor scheduling Simulation Benchmark testing Autonomous aerial vehicles approximation algorithms Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

Robust Path Planning Using Adaptive Reinforcement Learning in Simulation Environment

Robust Path Planning Using Adaptive Reinforcement Learning i...

引用

International Conference on Control, Automation and Information Sciences (ICCAIS)

作者： Daeyoel Kang Jun Young Yun Noheun Myeong Jongyoon Park Pileun Kim Department of Artificial Intelligence Korea Aerospace University Republic of Korea

ISBN: (数字)9798331542047

ISBN: (纸本)9798331542054

As autonomous driving gains widespread attention, extensive research is being conducted to enable robots to safely reach their destinations independently. Before the rise of advanced robotic artificial intelligence, much research focused on path planning for route exploration and planning. However, these methods have limitations in responding to dynamic environmental changes. To overcome this, efforts have been made to integrate reinforcement learning. This study proposes an adaptive path planning reinforcement learning algorithm, combining the efficiency of path planning with the adaptability of reinforcement learning to various environmental changes. The proposed method effectively integrates traditional path planning algorithms with reinforcement learning. We utilize the TD3 network as the backbone and apply the Artificial Potential Field algorithm, known for its robustness to obstacles. Experimental results demonstrate that our approach achieved approximately 14 points higher average rewards and similar maximum rewards compared to existing networks, indicating higher average success rates. Additionally, results specified above indicate faster or comparable convergence speeds relative to traditional networks.

关键词： Training Navigation Heuristic algorithms Decision making Reinforcement learning Robot sensing systems approximation algorithms Path planning Robustness Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

A New Approach for Approximating Directed Rooted Networks

arXiv

引用

arXiv 2024年

作者： Cohen, Sarel Kamma, Lior Niklanovits, Aikaterini The Academic College of Tel Aviv-Yaffo Israel Hasso Plattner Institute University of Potsdam Germany

We consider the k-outconnected directed Steiner tree problem (k-DST). Given a directed edge-weighted graph G = (V, E, w), where V = {r} ∪ S ∪ T, and an integer k, the goal is to find a minimum cost subgraph of G in which there are k edge-disjoint rt-paths for every terminal t ∈ T. The problem is known to be NP-Hard. Furthermore, the question on whether a polynomial time, subpolynomial approximation algorithm exists for k-DST was answered negatively by Grandoni et al. (2018), by proving an approximation hardness of Ω(|T|/log |T|) under NP≠ZPP. Inspired by modern day applications, we focus on developing efficient algorithms for k-DST in graphs where terminals have out-degree 0, and furthermore constitute the vast majority in the graph. We provide the first approximation algorithm for k-DST on such graphs, in which the approximation ratio depends (primarily) on the size of S. We present a randomized algorithm that finds a solution of weight at most O(k|S|log |T|) times the optimal weight, and with high probability runs in polynomial time. © 2024, CC BY.

关键词： approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

Efficient Hardware Implementation of Nonlinear Activation Function For Inference Model

Efficient Hardware Implementation of Nonlinear Activation Fu...

引用

International SoC Design Conference, ISOCC

作者： Snehit Chunarkar Samba Raju Chiluveru Electrical Department Indian Institute of Technolgy Roorkee EECE Indian Institute of Technology Dharwad

ISBN: (数字)9798350377088

ISBN: (纸本)9798350377095

This paper presents a novel piecewise linear (PWL) approximation method for designing a highly precise nonlinear activation function tailored for hardware implementation of inference models. It is developed as an Adaptive Step-Size-based Recursive Algorithm (ASRA) method, incorporating the maximum allowable error (ϵ) as an input parameter. PWL functions are realized with minimal computational overhead, utilizing only addition operations and coefficient memory, thus avoiding multiplications. With fewer resources, the proposed method allows for the accurate approximation of nonlinear functions. The hardware implementation uses a Synopsys Design Compiler with a TSMC 90-nm library. Performance comparison in terms of area, delay, and power consumption demonstrates the effectiveness of the proposed approach.

关键词： Semiconductor device modeling Adaptation models Power demand Accuracy Neural networks approximation algorithms Hardware Libraries Inference algorithms Delays

来源：评论

学校读者我要写书评

暂无评论

Prediction-based Coflow Scheduling

Prediction-based Coflow Scheduling

引用

International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt)

作者： Olivier Brun Balakrishna J. Prabhu Oumayma Haddaji LAAS-CNRS University of Toulouse CNRS INSA Toulouse France

ISBN: (数字)9783903176652

ISBN: (纸本)9798331508722

We explore the problem of minimizing the weighted average coflow completion time of coflows when flow sizes are unknown but unreliable predictions on them are available. We propose to use Sincronia, a 4-approximation algorithm when actual flow sizes are known, directly on the predictions. For this variant of Sincronia, we obtain a finite upper bound on the approximation ratio that combines unreliability on predictions and the weights of the coflows. On several numerical experiments, it is shown that this bound is too conservative, and that in practice Sincronia with predictions has a much better performance on average as well as in the worst-case.

关键词： Degradation Analytical models Upper bound Wireless networks Predictive models Prediction algorithms approximation algorithms Resource management Feeds Optimization

来源：评论

学校读者我要写书评

暂无评论

A Numerical Method for Prediction of Standard Partial Discharge Test Performance of an Open-Ventilated Dry-Type Transformer at Design Stage

A Numerical Method for Prediction of Standard Partial Discha...

引用

Electrical Machines and Drives (ICEMD), International Conference on

作者： Mohammad Qumi Mahdi Kazemiun Mehdi Vakilian Dept. of Electrical Engineering Sharif University of Technology Tehran Iran

ISBN: (数字)9798331507510

ISBN: (纸本)9798331507527

Partial discharge in an (open ventilated dry type) OVDT sub-transmission transformer with 30 MVA 63kV/20kV rating has been studied in this work. Nomex has been used in this transformer in the form of insulation paper. The goal of this paper is to present a method to predict the result of the standard partial discharge test on a dry-type transformer based on its design parameters. The importance of such a method is in the prevention of manufacturing a power transformer that might be rejected due to its lack of endurance in the partial discharge test which can lead to cost savings. In this method, important areas in transformer winding insulation design are identified and in turn, effective factors on electric field magnitude in these areas are determined. Then, using a Monte Carlo-based method, a numeric algorithm for calculating the electric field magnitude in these areas is developed. Finally, based on the electric field magnitude distribution in these important areas it will be determined whether the transformer insulation would pass the requirements of the IEC standard partial discharge test or not. A comparison of the results obtained using the algorithm against the finite element simulation results showed an error of less than 1%.

关键词： Partial discharges Insulation Monte Carlo methods Simulation Windings approximation algorithms Finite element analysis Electric fields Standards Power transformer insulation

来源：评论

学校读者我要写书评

暂无评论

Deterministic Algorithm and Faster Algorithm for Submodular Maximization Subject to a Matroid Constraint

Deterministic Algorithm and Faster Algorithm for Submodular ...

引用

Annual IEEE Symposium on Foundations of Computer Science

作者： Niv Buchbinder Moran Feldman Department of Statistics and Operations Research Tel Aviv University Tel-Aviv Israel Department of Computer Science University of Haifa Haifa Israel

ISBN: (数字)9798331516741

ISBN: (纸本)9798331516758

We study the problem of maximizing a monotone submodular function subject to a matroid constraint, and present for it a deterministic non-oblivious local search algorithm that has an approximation guarantee of $1-1/e-\epsilon$ (for any $\epsilon > 0$ ) and query complexity of $\tilde{O}_{\epsilon}(nr)$ , where $n$ is the size of the ground set and $r$ is the rank of the matroid. Our algorithm vastly improves over the previous state-of-the-art 0.5008-approximation deterministic algorithm, and in fact, shows that there is no separation between the approximation guarantees that can be obtained by deterministic and randomized algorithms for the problem considered. The query complexity of our algorithm can be improved to $\tilde{O}_{\epsilon}(n+\hat{r}\sqrt{{n}})$ using randomization, which is nearly-linear for $r=O(\sqrt{n})$ , and is always at least as good as the previous state-of-the-art algorithms.

关键词： Computer science approximation algorithms Search problems Complexity theory

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：