检索结果-内蒙古大学图书馆

Proceedings of the 41st International Conference on Machine Learning

作者： Laixi Shi Eric Mazumdar Yuejie Chi Adam Wierman Department of Computing Mathematical Sciences California Institute of Technology CA Department of Electrical and Computer Engineering Carnegie Mellon University Pittsburgh PA

To overcome the sim-to-real gap in reinforcement learning (RL), learned policies must maintain robustness against environmental uncertainties. While robust RL has been widely studied in single-agent regimes, in multi-agent environments, the problem remains understudied-- despite the fact that the problems posed by environmental uncertainties are often exacerbated by strategic interactions. This work focuses on learning in distributionally robust Markov games (RMGs), a robust variant of standard Markov games, wherein each agent aims to learn a policy that maximizes its own worst-case performance when the deployed environment deviates within its own prescribed uncertainty set. This results in a set of robust equilibrium strategies for all agents that align with classic notions of game-theoretic equilibria. Assuming a non-adaptive sampling mechanism from a generative model, we propose a sample-efficient model-based algorithm (DRNVI) with finite-sample complexity guarantees for learning robust variants of various notions of game-theoretic equilibria. We also establish an information-theoretic lower bound for solving RMGs, which confirms the near-optimal sample complexity of DR-NVI with respect to problem-dependent factors such as the size of the state space, the target accuracy, and the horizon length.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Limitations of nearest-neighbor quantum networks

引用

Physical Review A 2022年第1期106卷 L010401-L010401页

作者： F. Hahn A. Dahlberg J. Eisert A. Pappa Dahlem Center for Complex Quantum Systems Freie Universität Berlin 14195 Berlin Germany Department of Electrical Engineering and Computer Science Technische Universität Berlin 10587 Berlin Germany QuTech TU Delft 2628CJ Delft The Netherlands Fraunhofer Heinrich Hertz Institute 10587 Berlin Germany Department of Electrical Engineering and Computer Science Technische Universität Berlin 10587 Berlin Germany Fraunhofer Institute for Open Communication Systems-FOKUS 10589 Berlin Germany

quantum communication research has in recent years shifted to include multipartite networks for which questions of quantum network routing naturally emerge. To understand the potential for multipartite routing, we focus on the most promising architectures for future quantum networks—those connecting nodes close to each other. Nearest-neighbor networks, such as rings, lines, and grids, have been studied under different communication scenarios to facilitate the sharing of quantum resources especially in the presence of bottlenecks. We analyze the potential of nearest-neighbor entangling gate quantum networks and identify some serious limitations by demonstrating that rings and lines cannot overcome common bottleneck communication problems.

关键词： Entanglement manipulation quantum communication quantum entanglement quantum information theory quantum networks quantum repeaters

来源：评论

学校读者我要写书评

暂无评论

Tailoring Fusion-Based Photonic quantum computing Schemes to quantum Emitters

引用

PRX quantum 2025年第2期6卷 020304-020304页

作者： Ming Lai Chan Thomas J. Bell Love A. Pettersson Susan X. Chen Patrick Yard Anders S. Sørensen Stefano Paesani Sparrow Quantum Blegdamsvej 104A DK-2100 Copenhagen Ø Denmark Quantum Engineering Technology Labs H. H. Wills Physics Laboratory and Department of Electrical and Electronic Engineering University of Bristol Bristol BS8 1FD United Kingdom Quantum Engineering Centre for Doctoral Training University of Bristol Bristol United Kingdom NNF Quantum Computing Programme Niels Bohr Institute University of Copenhagen Blegdamsvej 17 DK-2100 Copenhagen Ø Denmark

Fusion-based quantum computation is a promising quantum computing model where small-sized photonic resource states are simultaneously entangled and measured by fusion gates. Such operations can be readily implemented with scalable photonic hardware: resource states can be deterministically generated by quantum emitters and fusions require only shallow linear-optical circuits. Here, we propose fusion-based architectures tailored to the capabilities and noise models in quantum emitters. We show that high tolerance to dominant physical error mechanisms can be achieved, with fault-tolerance thresholds of 8% for photon loss, 4% for photon distinguishability between emitters, and spin noise thresholds well above memory-induced errors for typical spin-photon interfaces. Our construction and analysis provide guidelines for the development of photonic quantum hardware targeting fault-tolerant applications with quantum emitters.

关键词： quantum computers

来源：评论

学校读者我要写书评

暂无评论

Landscape approximation of low-energy solutions to binary optimization problems

引用

Physical Review A 2024年第1期109卷 012433-012433页

作者： Benjamin Y. L. Tan Beng Yee Gan Daniel Leykam Dimitris G. Angelakis Centre for Quantum Technologies National University of Singapore 3 Science Drive 2 Singapore 117543 Singapore School of Electrical and Computer Engineering Technical University of Crete Chania 73100 Greece AngelQ Quantum Computing 531A Upper Cross Street 04-95 Hong Lim Complex Singapore 051531 Singapore

We show how the localization landscape, originally introduced to bound low-energy eigenstates of disordered wave media and many-body quantum systems, can form the basis for hardware-efficient quantum algorithms for solving binary optimization problems. Many binary optimization problems can be cast as finding low-energy eigenstates of Ising Hamiltonians. First, we apply specific perturbations to the Ising Hamiltonian such that the low-energy modes are bounded by the localization landscape. Next, we demonstrate how a variational method can be used to prepare and sample from the peaks of the localization landscape. Numerical simulations of problems of up to ten binary variables show that the localization landscape-based sampling can outperform quantum approximate optimization algorithm (QAOA) circuits of similar depth, as measured in terms of the probability of sampling the exact ground state.

关键词： quantum algorithms quantum algorithms & computation quantum computation

来源：评论

学校读者我要写书评

暂无评论

Prototype Guided Personalized Federated Intrusion Detection System 10

Prototype Guided Personalized Federated Intrusion Detection ...

引用

10th IEEE Smart World Congress, SWC 2024

作者： Cheng, Long Yan, Huiru Zhou, Hanlin Wang, Ying Tang, Haichuan Fang, Fang North China Electric Power University State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources School of Control and Computer Engineering Beijing China Institute of Computing Technology Chinese Academy of Sciences Beijing China Ai Lab of Crrc Academy Beijing China

ISBN: (纸本)9798331520861

Preventing network attacks and protecting user privacy are consistently hot research topics in the Internet of Things (IoT) and edge computing fields. Recent advancements in Federated Learning (FL) have shown promise in addressing these challenges. FL allows various clients to collaboratively build an intrusion detection system (IDS) without sharing their private data. However, most existing methods train a single intrusion detection model for all clients and assume that the training and test data distribution are identical, leading to unsatisfactory detection accuracy and generalization abilities in practice. To overcome these challenges, this paper introduces a prototype-guided personalized FL approach named PG-FedIDS. We propose two novel mechanisms within this method. Firstly, we utilize class prototypes as auxiliary information carriers to generate personalized models for each client, rather than generating a single global model as in previous works. Secondly, we propose a prototype-guided ensemble learning strategy, which can leverage the global knowledge in prototypes to enhance detection accuracy and generalization abilities for each client. We conduct extensive experiments on two benchmark datasets with different evaluation test settings. The results demonstrate that our PG-FedIDS achieves promising detection accuracy and consistently outperforms other FL baselines. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Federated offline reinforcement learning: collaborative single-policy coverage suffices 24

Federated offline reinforcement learning: collaborative sing...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Jiin Woo Laixi Shi Gauri Joshi Yuejie Chi Department of Electrical and Computer Engineering Carnegie Mellon University Pittsburgh PA Department of Computing Mathematical Sciences California Institute of Technology CA

Offline reinforcement learning (RL), which seeks to learn an optimal policy using offline data, has garnered significant interest due to its potential in critical applications where online data collection is infeasible or expensive. This work explores the benefit of federated learning for offline RL, aiming at collaboratively leveraging offline datasets at multiple agents. Focusing on finite-horizon episodic tabular Markov decision processes (MDPs), we design FedLCB-Q, a variant of the popular model-free Q-learning algorithm tailored for federated offline RL. FedLCB-Q updates local Q-functions at agents with novel learning rate schedules and aggregates them at a central server using importance averaging and a carefully designed pessimistic penalty term. Our sample complexity analysis reveals that, with appropriately chosen parameters and synchronization schedules, FedLCB-Q achieves linear speedup in terms of the number of agents without requiring high-quality datasets at individual agents, as long as the local datasets collectively cover the state-action space visited by the optimal policy, highlighting the power of collaboration in the federated setting. In fact, the sample complexity almost matches that of the single-agent counterpart, as if all the data are stored at a central location, up to polynomial factors of the horizon length. Furthermore, FedLCB-Q is communication-efficient, where the number of communication rounds is only linear with respect to the horizon length up to logarithmic factors.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficiency of Wireless Communication in Cybersecurity for Data Security 10

Efficiency of Wireless Communication in Cybersecurity for Da...

引用

10th IEEE Uttar Pradesh Section International Conference on electrical, Electronics and computer engineering, UPCON 2023

作者： Shashank, R. Krishna, Konda Hari Takore, Tamirat Tagesse Singla, Atul Veena, Ch Tiwari, Mohit School of Engineering Sr Universtity Department of ECE Telangana Warangal India School of Computing Mohan Babu University Dept. of Cse Tirupati India Wachemo University Department of Electrical and Computer Engineering Ethiopia Ethiopia Lovely Professional University Phagwara India Institute of Aeronautical Engineering Hyderabad India Bharati Vidyapeeth's College of Engineering Department of Computer Science and Engineering Delhi India

ISBN: (纸本)9798350382471

This Article's aim is to analyse the wireless communication in Cybersecurity. The topic of this paper provides readers with a better grasp of the safety guidelines and measures that are currently on the market, as well as a quick review of the advantages and disadvantages of each kind of security approach. The potential of the system to manage, protect, and distribute confidential data is determined by protection in the world of computers. © 2023 IEEE.

关键词： Cybersecurity data security wireless communication

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning for Accurate Anomaly Detection in Multi-Cloud Environment 2

Deep Reinforcement Learning for Accurate Anomaly Detection i...

引用

2nd IEEE Global Conference on Communications and Information Technologies, GCCIT 2024

作者： Chaudhary, Anisha Nalini Sunil, M. Tiwaskar, S.A. Vijayalakshmi, V.J. Sharma, Sorabh Senthurya, S. Quantum University Research Center Quantum University Uttarakhand Roorkee247667 India School of Mangement - Pg Management Karnataka Bangalore India Vishwakarma Institute of Technology Pune India Karpagam Academy of Higher Education Department of Electrical and Electronics Engineering Coimbatore641021 India Chitkara University Centre of Research Impact and Outcome Punjab Rajpura140417 India Prince Shri Venkateshwara Padmavathy Engineering College Department of Computer Science and Engineering Chennai127 India

ISBN: (纸本)9798350388916

Intrusion Detection System (IDS) is used for identifying complicated Cloud assaults. A cooperative IDS may use an aggregation technique to decide on suspicious intrusions. Aggregation methods and IDS feedback slow the process. These restrictions make real-time cooperative IDS choices useless and unsustainable. We suggest a machine learning-based collaborative IDS that uses past feedback data to make proactive decisions to address these issues. DRL-based multi-cloud IDS was suggested. The suggested approach uses a DE noising Auto encoder (DA) to create a deep neural network. DA can rebuild IDS feedback from incomplete feedback. The DA network uses Q-learning to continuously learn features. This lets us respond on suspected incursions without IDS input. The GPU-enabled Tensor Flow model was tested on a real-world dataset. Our model achieves 95% detection accuracy in experiments. © 2024 IEEE.

关键词： Anomaly Detection Cloud Environment Deep Reinforcement DRL Effective Anomaly Identification and IDS Multi-Cloud Environment

来源：评论

学校读者我要写书评

暂无评论

SRC: Sustainable Reactive computing for Battery-free Edge Intelligence

SRC: Sustainable Reactive Computing for Battery-free Edge In...

引用

International Green computing Conference

作者： Sepehr Tabrizchi Nedasadat Taheri Justin Feng Nader Sehatbakhsh David Z. Pan Arman Roohi Department of Electrical and Computer Engineering University of Illinois Chicago Chicago IL USA School of Computing University of Nebraska-Lincoln Lincoln NE USA Department of Electrical and Computer Engineering New Jersey Institute of Technology Newark NJ USA Department of Electrical and Computer Engineering University of Texas Austin Austin

ISBN: (数字)9798331507862

ISBN: (纸本)9798331507879

This paper proposes SRC, a novel framework for efficient and reliable inference on battery-free smart Internet of Things (IoT) devices. SRC supports various configurations that follow reactive configuration while using the innovative state machine and a safe threshold mechanism to proactively halt operations, reducing store/load operations by up to 75%. It strategically stores essential convolutional neural network (CNN) data (layer, kernel, etc.) to optimize input/output feature map management. This reactive design allows seamless task resumption across power cycles, ensuring continuity in unpredictable energy environments. Experiments show significant gains, with SRC achieving on average ~ 81.85% reduction in read/write operations and approximately 57.18% improvement in sensing compared to conventional reactive methods based on the intermittent Energy Trace 1.

关键词： Green products Sensors Internet of Things Convolutional neural networks Reliability Kernel

来源：评论

学校读者我要写书评

暂无评论

EFFICIENT REINFORCEMENT LEARNING FOR GLOBAL DECISION MAKING IN THE PRESENCE OF LOCAL AGENTS AT SCALE

arXiv

引用

arXiv 2024年

作者： Anand, Emile Qu, Guannan Computing and Mathematical Sciences California Institute of Technology United States Department of Electrical and Computer Engineering Carnegie Mellon University United States

We study reinforcement learning for global decision-making in the presence of local agents, where the global decision-maker makes decisions affecting all local agents, and the objective is to learn a policy that maximizes the joint rewards of all the agents. Such problems find many applications, e.g. demand response, EV charging, queueing, etc. In this setting, scalability has been a long-standing challenge due to the size of the state space which can be exponential in the number of agents. This work proposes the SUBSAMPLE-Q algorithm where the global agent subsamples k ≤ n local agents to compute a policy in time that is polynomial in k. We show that this learned policy converges to the optimal policy in the order of Õ(1/√k + ϵk,m) as the number of sub-sampled agents k increases, where ϵk,m is the Bellman noise. Finally, we validate the theory through numerical simulations in a demand-response setting and a queueing setting. © 2024, CC BY-SA.

关键词： Reinforcement learning

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：