检索结果-内蒙古大学图书馆

Two-Player Multiagent Graphical Games with Reinforcement Learning*

学校读者我要写书评

暂无评论

Two-Player Multiagent Graphical Games with Reinforcement Lea...

Industrial Cyber-Physical systems (ICPS)

作者： Bosen Lian Jiacheng Wu Electrical and Computer Engineering Department Auburn University Auburn AL USA State Key Laboratory of Industrial Control Technology Zhejiang University Hangzhou China

ISBN: (数字)9798350363012

ISBN: (纸本)9798350363029

This paper studies the synchronization problem of two-player multiagent systems through reinforcement learning methods. A Nash-minmax strategy is formulated, where the interactions of two players in the same agent are non-zero-sum, while interactions of players between agents are zero-sum games. We propose an offline model-based reinforcement learning algorithm to identify Nash solutions for players within each agent, as well as the worst control solutions for players in neighboring antagonistic agents. On this basis, a data-driven off-policy algorithm is provided to alleviate the requirement for accurate system dynamics in the offline algorithm. Besides, the convergence of the proposed algorithms is analyzed. Finally, simulation results verify the effectiveness of the designed algorithms.

关键词： System dynamics Heuristic algorithms Simulation Games Reinforcement learning Cyber-physical systems Synchronization

Explainable-AI-based two-stage solution for WSN object localization using zero-touch mobile transceivers

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第7期67卷 27-45页

作者： Kai FANG Junxin CHEN Han ZHU Thippa Reddy GADEKALLU Xiaoping WU Wei WANG School of Mathematics and Computer Science Zhejiang A&F University School of Software Dalian University of Technology Faculty of Applied Sciences Macao Polytechnic University Department of Electrical and Computer Engineering Lebanese American University School of Information Engineering Huzhou University Guangdong-Hong Kong-Macao Joint Laboratory for Emotional Intelligence and Pervasive Computing Artificial Intelligence Research Institute Shenzhen MSU-BIT University

Artificial intelligence technology is widely used in the field of wireless sensor networks(WSN).Due to its inexplicability, the interference factors in the process of WSN object localization cannot be effectively eliminated. In this paper, an explainable-AI-based two-stage solution is proposed for WSN object localization. In this solution, mobile transceivers are used to enlarge the positioning range and eliminate the blind area for object localization. The motion parameters of transceivers are considered to be unavailable,and the localization problem is highly nonlinear with respect to the unknown parameters. To address this,an explainable AI model is proposed to solve the localization problem. Since the relationship among the variables is difficult to fully include in the first-stage traditional model, we develop a two-stage explainable AI solution for this localization problem. The two-stage solution is actually a comprehensive consideration of the relationship between variables. The solution can continue to use the constraints unused in the firststage during the second-stage, thereby improving the performance of the solution. Therefore, the two-stage solution has stronger robustness compared to the closed-form solution. Experimental results show that the performance of both the two-stage solution and the traditional solution will be affected by numerical changes in unknown parameters. However, the two-stage solution performs better than the traditional solution, especially with a small number of mobile transceivers and sensors or in the presence of high noise. Furthermore,we have also verified the feasibility of the proposed explainable-AI-based two-stage solution.

关键词： explainable AI object localization semidefinite relaxation mobile transceiver two-stage solution closed-form solution

Adversarial Attacks on Deep Neural Networks using Generative Perturbation Networks

学校读者我要写书评

暂无评论

Adversarial Attacks on Deep Neural Networks using Generative...

International Conference on Ubiquitous and Future Networks, ICUFN

作者： Sudharman K. Jayaweera Department of Electrical and Computer Engineering Communications and Information Sciences Laboratory (CISL) University of New Mexico Albuquerque NM USA

A new deep learning (DL) network called Generative Perturbation Network (GPN) is proposed. The GPNs are capable of learning to modify inputs to deep neural network (DNN) models trained to classify images in imperceptible ways leading to misclassification. Unlike previous approaches to generate such adversarial samples to DNN classifiers, the proposed GPN does not need to know the architecture of the neural network that it is trying to attack nor have access to the data used to trained it. It is shown that with just being able to observe the final output labels from the trained classifier to any given input image, the GPNs are able to learn to minimally perturb the images to achieve misclassification. Simulation results show that a proposed GPN can easily degrade the 98.65% accuracy of a trained CNN on the MNIST hand-written digit dataset down to about 52%. Simulation results also show that different regularizations can be used in the generator loss function to achieve various visually desirable characteristics in the generated adversarially perturbed images.

关键词：

Projection-Free Methods for Stochastic Simple Bilevel Optimization with Convex Lower-level Problem

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Cao, Jincheng Jiang, Ruichen Abolfazli, Nazanin Hamedani, Erfan Yazdandoost Mokhtari, Aryan Department of Electrical and Computer Engineering The University of Texas AustinTX United States Department of Systems and Industrial Engineering The University of Arizona TucsonAZ United States

In this paper, we study a class of stochastic bilevel optimization problems, also known as stochastic simple bilevel optimization, where we minimize a smooth stochastic objective function over the optimal solution set of another stochastic convex optimization problem. We introduce novel stochastic bilevel optimization methods that locally approximate the solution set of the lower-level problem via a stochastic cutting plane, and then run a conditional gradient update with variance reduction techniques to control the error induced by using stochastic gradients. For the case that the upper-level function is convex, our method requires Õ(max{1/Ε2f, 1/Ε2g}) stochastic oracle queries to obtain a solution that is Εf-optimal for the upper-level and Εgoptimal for the lower-level. This guarantee improves the previous best-known complexity of O(max{1/Ε4f, 1/Ε4g}). Moreover, for the case that the upper-level function is non-convex, our method requires at most Õ(max{1/Ε3f, 1/Ε3g}) stochastic oracle queries to find an (Εf, Εg)stationary point. In the finite-sum setting, we show that the number of stochastic oracle calls required by our method are Õ(√n/Ε) and Õ(√n/Ε2) for the convex and non-convex settings, respectively, where Ε = min{Εf, Εg}. Copyright © 2023, The Authors. All rights reserved.

关键词： Stochastic systems

BitMoD: Bit-serial Mixture-of-Datatype LLM Acceleration

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Yuzong AbouElhamayed, Ahmed F. Dai, Xilai Wang, Yang Andronic, Marta Constantinides, George A. Abdelfattah, Mohamed S. Computer Systems Lab Cornell University United States Systems and Networking Research Group Microsoft Research United States Department of Electrical and Electronic Engineering Imperial College London United Kingdom

Large language models (LLMs) have demonstrated remarkable performance across various machine learning tasks. Yet the substantial memory footprint of LLMs significantly hinders their deployment. In this paper, we improve the accessibility of LLMs through BitMoD1, an algorithm-hardware co-design solution that enables efficient LLM acceleration at low weight precision. On the algorithm side, BitMoD introduces fine-grained data type adaptation that uses a different numerical data type to quantize a group of (e.g., 128) weights. Through the careful design of these new data types, BitMoD is able to quantize LLM weights to very low precision (e.g., 4 bits and 3 bits) while maintaining high accuracy. On the hardware side, BitMoD employs a bit-serial processing element to easily support multiple numerical precisions and data types;our hardware design includes two key innovations: First, it employs a unified representation to process different weight data types, thus reducing the hardware cost. Second, it adopts a bit-serial dequantization unit to rescale the per-group partial sum with minimal hardware overhead. Our evaluation on six representative LLMs demonstrates that BitMoD significantly outperforms state-of-the-art LLM quantization and acceleration methods. For discriminative tasks, BitMoD can quantize LLM weights to 4-bit with © 2024, CC BY-NC-ND.

关键词：

Hierarchical Micro-Segmentations for Zero-Trust Services via Large Language Model (LLM)-enhanced Graph Diffusion

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Yinqiu Liu, Guangyuan Du, Hongyang Niyato, Dusit Kang, Jiawen Xiong, Zehui In Kim, Dong Shen, Xuemin College of Computing and Data Science Nanyang Technological University Singapore School of Automation Guangdong University of Technology China Pillar of Information Systems Technology and Design Singapore University of Technology and Design Singapore Department of Electrical and Computer Engineering Sungkyunkwan University Korea Republic of Department of Electrical and Computer Engineering University of Waterloo Canada

—In the rapidly evolving Next-Generation Networking (NGN) era, the adoption of zero-trust architectures has become increasingly crucial to protect security. However, provisioning zero-trust services in NGNs poses significant challenges, primarily due to the environmental complexity and dynamics. Motivated by these challenges, this paper explores efficient zero-trust service provisioning using hierarchical micro-segmentations. Specifically, we model zero-trust networks via hierarchical graphs, thereby jointly considering the resource- and trust-level features to optimize service efficiency. We organize such zero-trust networks through micro-segmentations, which support granular zero-trust policies efficiently. To generate the optimal micro-segmentation, we present the Large Language Model-Enhanced Graph Diffusion (LEGD) algorithm, which leverages the diffusion process to realize a high-quality generation paradigm. Additionally, we utilize policy boosting and Large Language Models (LLM) to enable LEGD to optimize the generation policy and understand complicated graphical features. Moreover, realizing the unique trustworthiness updates or service upgrades in zero-trust NGN, we further present LEGD-Adaptive Maintenance (LEGD-AM), providing an adaptive way to perform task-oriented fine-tuning on LEGD. Extensive experiments demonstrate that the proposed LEGD achieves 90% higher efficiency in provisioning services compared with other baselines. Moreover, the LEGD-AM can reduce the service outage time by over 50%. © 2024, CC BY.

关键词： Diffusion

Differentially Private Online Federated Learning with Correlated Noise

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Jiaojiao Zhu, Linglingzhi Johansson, Mikael Division of Decision and Control Systems School of Electrical Engineering and Computer Science KTH Royal Institute of Technology StockholmSE-100 44 Sweden Department of Systems Engineering and Engineering Management The Chinese University of Hong Kong Hong Kong

We introduce a novel differentially private algorithm for online federated learning that employs temporally correlated noise to enhance utility while ensuring privacy of continuously released models. To address challenges posed by DP noise and local updates with streaming non-iid data, we develop a perturbed iterate analysis to control the impact of the DP noise on the utility. Moreover, we demonstrate how the drift errors from local updates can be effectively managed under a quasi-strong convexity condition. Subject to an (ϵ, δ)DP budget, we establish a dynamic regret bound over the entire time horizon, quantifying the impact of key parameters and the intensity of changes in dynamic environments. Numerical experiments confirm the efficacy of the proposed algorithm. © 2024, CC BY.

关键词： Federated learning

Distributed and Rate-Adaptive Feature Compression

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Deshmukh, Aditya Veeravalli, Venugopal V. Verma, Gunjan Department of Electrical and Computer Engineering University of Illinois Urbana-Champaign ChampaignIL United States Army Research Laboratory AdelphiMD United States

We study the problem of distributed and rateadaptive feature compression for linear regression. A set of distributed sensors collect disjoint features of regressor data. A fusion center is assumed to contain a pretrained linear regression model, trained on a dataset of the entire uncompressed data. At inference time, the sensors compress their observations and send them to the fusion center through communication-constrained channels, whose rates can change with time. Our goal is to design a feature compression scheme that can adapt to the varying communication constraints, while maximizing the inference performance at the fusion center. We first obtain the form of optimal quantizers assuming knowledge of underlying regressor data distribution. Under a practically reasonable approximation, we then propose a distributed compression scheme which works by quantizing a one-dimensional projection of the sensor *** also propose a simple adaptive scheme for handling changes in communication constraints. We demonstrate the effectiveness of the distributed adaptive compression scheme through simulated experiments. Copyright © 2024, The Authors. All rights reserved.

关键词： Machine learning

Autonomous Patrol and Threat Detection Through Integrated Mapping and computer Vision

学校读者我要写书评

暂无评论

Autonomous Patrol and Threat Detection Through Integrated Ma...

IEEE International Conference on Electro-Information Technology

作者： Robert Soler Alae Moudni Gabriel Roskowski Xinrui Yu Mikhail Gormov Jafar Saniie Department of Electrical and Computer Engineering Embedded Computing and Signal Processing (ECASP) Research Laboratory Illinois Institute of Technology Chicago IL U.S.A.

ISBN: (数字)9798350330649

ISBN: (纸本)9798350330656

This paper presents the creation of an innovative autonomous security robot designed to perform security functions with efficiency and reliability. The robot boasts mapping capabilities, which it utilizes to facilitate autonomous patrol in designated areas. Its primary operations involve the use of computer vision to detect violence, identify weapons and dangerous items, and recognize individuals. Critical incidents are met with an immediate alarm and the subsequent transmission of data to a central security server, which then generates comprehensive reports displayed through a web application for security personnel. The application itself features remote control of the robot, incident report management, status updates, and incident analytics. The robot demonstrates substantial real-world application potential, particularly in crowded environments where it could outperform conventional surveillance. The project combines concepts of engineering, computer science, and cybersecurity, functioning per design but with considerable potential for future refinement and expansion, embodying the concept of an evolving technological solution.

关键词： computer science computer vision Technological innovation Surveillance Weapons Reliability engineering Threat assessment