检索结果-内蒙古大学图书馆

Federated Linear Bandit Learning via Over-the-air Computation

学校读者我要写书评

暂无评论

Federated Linear Bandit Learning via Over-the-air Computatio...

IEEE Conference on Global Communications (GLOBECOM)

作者： Jiali Wang Yuning Jiang Xin Liu Ting Wang Yuanming Shi MoE Engineering Research Center of Software/Hardware Co-design Technology and Application Shanghai Key Lab. of Trustworthy Computing East China Normal University China School of Information Science and Technology (SIST) ShanghaiTech University China Automatic Control Laboratory Ecole Polytechnique Fédérale de Lausanne (EPFL) Switzerland

In this paper, we investigate federated contextual linear bandit learning within a wireless system that comprises a server and multiple devices. Each device interacts with the environment, selects an action based on the received reward, and sends model updates to the server. The primary objective is to minimize cumulative regret across all devices within a finite time horizon. To reduce the communication overhead, devices communicate with the server via over-the-air computation (AirComp) over noisy fading channels, where the channel noise may distort the signals. In this context, we propose a customized federated linear bandits scheme, where each device transmits an analog signal, and the server receives a superposition of these signals distorted by channel noise. A rigorous mathematical analysis is conducted to determine the regret bound of the proposed scheme. Both theoretical analysis and numerical experiments demonstrate the competitive performance of our proposed scheme in terms of regret bounds in various settings.

关键词：

Preserving earlier knowledge in continual learning with the help of all previous feature extractors

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Li, Zhuoyun Zhong, Changhong Liu, Sijia Wang, Ruixuan Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University China Key Laboratory of Machine Intelligence and Advanced Computing MOE Guangzhou510000 China

Continual learning of new knowledge over time is one desirable capability for intelligent systems to recognize more and more classes of objects. Without or with very limited amount of old data stored, an intelligent system often catastrophically forgets previously learned old knowledge when learning new knowledge. Recently, various approaches have been proposed to alleviate the catastrophic forgetting issue. However, old knowledge learned earlier is commonly less preserved than that learned more recently. In order to reduce the forgetting of particularly earlier learned old knowledge and improve the overall continual learning performance, we propose a simple yet effective fusion mechanism by including all the previously learned feature extractors into the intelligent model. In addition, a new feature extractor is included to the model when learning a new set of classes each time, and a feature extractor pruning is also applied to prevent the whole model size from growing rapidly. Experiments on multiple classification tasks show that the proposed approach can effectively reduce the forgetting of old knowledge, achieving state-of-the-art continual learning performance. Copyright © 2021, The Authors. All rights reserved.

关键词： Intelligent systems

Linear uncertainty quantification of graphical model inference 24

学校读者我要写书评

暂无评论

Linear uncertainty quantification of graphical model inferen...

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Chenghua Guo Han Yu Jiaxin Liu Chao Chen Qi Li Sihong Xie Xi Zhang Key Laboratory of Trustworthy Distributed Computing and Service (MoE) Beijing University of Posts and Telecommunications China College of Artificial Intelligence Beijing University of Posts and Telecommunications China Department of Computer Science and Engineering Lehigh University School of Computer Science and Technology Harbin Institute of Technology (Shenzhen) China Department of Computer Science Iowa State University AI Thrust The Hong Kong University of Science and Technology (Guangzhou) China

ISBN: (纸本)9798331314385

Uncertainty Quantification (UQ) is vital for decision makers as it offers insights into the potential reliability of data and model, enabling more informed and risk-aware decision-making. Graphical models, capable of representing data with complex dependencies, are widely used across domains. Existing sampling-based UQ methods are unbiased but cannot guarantee convergence and are time-consuming on large-scale graphs. There are fast UQ methods for graphical models with closed-form solutions and convergence guarantee but with uncertainty underestimation. We propose LinUProp, a UQ method that utilizes a novel linear propagation of uncertainty to model uncertainty among related nodes additively instead of multiplicatively, to offer linear scalability, guaranteed convergence, and closed-form solutions without underestimating uncertainty. Theoretically, we decompose the expected prediction error of the graphical model and prove that the uncertainty computed by LinUProp is the generalized variance component of the decomposition. Experimentally, we demonstrate that LinUProp is consistent with the sampling-based method but with linear scalability and fast convergence. Moreover, LinUProp outperforms competitors in uncertainty-based active learning on four real-world graph datasets, achieving higher accuracy with a lower labeling budget.

关键词：

EFFECT OF RANDOM NOISES ON PATHWISE SOLUTIONS TO THE HIGH-DIMENSIONAL MODIFIED EULER-POINCARÉ SYSTEM

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhang, Lei School of Mathematics and Statistics Hubei Key Laboratory of Engineering Modeling and Scientific Computing Huazhong University of Science and Technology Hubei Wuhan430074 China

In this paper, we study the Cauchy problem for the stochastically perturbed high-dimensional modified Euler-Poincaré system (MEP2) on the torus Td, d ≥ 1. We first establish a local well-posedness framework in the sense of Hadamard for the MEP2 driven by general nonlinear multiplicative noises. Then two kinds of global existence and uniqueness results are demonstrated: One indicates that the MEP2 perturbed by nonlocal-type random noises with proper intensity admits a unique large global strong solution;The other one infers that, if the initial data is sufficiently small, then the MEP2 perturbed by linear multiplicative noise has a unique global solution with high probability. In the case of one dimension, we find that the stochastic MEP2 will break down in finite time when the initial data meets appropriate shape condition. Copyright © 2021, The Authors. All rights reserved.

关键词： Stochastic systems

Federated Reinforcement Learning for Electric Vehicles Charging Control on Distribution Networks

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Qian, Junkai Jiang, Yuning Liu, Xin Wang, Qiong Wang, Ting Shi, Yuanming Chen, Wei The MoE Engineering Research Center of Software/Hardware Co-design Technology and Application The Shanghai Key Lab. of Trustworthy Computing East China Normal University China The Automatic Control Laboratory EPFL Switzerland The School of Information Science and Technology ShanghaiTech University China The State Grid Beijing Electric Power Company China The Department of Electronic Engineering Tsinghua University China

With the growing popularity of electric vehicles (EVs), maintaining power grid stability has become a significant challenge. To address this issue, EV charging control strategies have been developed to manage the switch between vehicle-to-grid (V2G) and grid-to-vehicle (G2V) modes for EVs. In this context, multi-agent deep reinforcement learning (MADRL) has proven its effectiveness in EV charging control. However, existing MADRL-based approaches fail to consider the natural power flow of EV charging/discharging in the distribution network and ignore driver privacy. To deal with these problems, this paper proposes a novel approach that combines multi-EV charging/discharging with a radial distribution network (RDN) operating under optimal power flow (OPF) to distribute power flow in real time. A mathematical model is developed to describe the RDN load. The EV charging control problem is formulated as a Markov Decision Process (MDP) to find an optimal charging control strategy that balances V2G profits, RDN load, and driver anxiety. To effectively learn the optimal EV charging control strategy, a federated deep reinforcement learning algorithm named FedSAC is further proposed. Comprehensive simulation results demonstrate the effectiveness and superiority of our proposed algorithm in terms of the diversity of the charging control strategy, the power fluctuations on RDN, the convergence efficiency, and the generalization ability. © 2023, CC BY-NC-SA.

关键词： Reinforcement learning

Decentralized Optimization Over the Stiefel Manifold by an Approximate Augmented Lagrangian Function

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Wang, Lei Liu, Xin State Key Laboratory of Scientific and Engineering Computing Academy of Mathematics and Systems Science Chinese Academy of Sciences University of Chinese Academy of Sciences China

In this paper, we focus on the decentralized optimization problem over the Stiefel manifold, which is defined on a connected network of d agents. The objective is an average of d local functions, and each function is privately held by an agent and encodes its data. The agents can only communicate with their neighbors in a collaborative effort to solve this problem. In existing methods, multiple rounds of communications are required to guarantee the convergence, giving rise to high communication costs. In contrast, this paper proposes a decentralized algorithm, called DESTINY, which only invokes a single round of communications per iteration. DESTINY combines gradient tracking techniques with a novel approximate augmented Lagrangian function. The global convergence to stationary points is rigorously established. Comprehensive numerical experiments demonstrate that DESTINY has a strong potential to deliver a cutting-edge performance in solving a variety of testing problems. Copyright © 2021, The Authors. All rights reserved.

关键词： Iterative methods

Moiré cavity quantum electrodynamics

学校读者我要写书评

暂无评论

Science Advances 2025年第21期11卷 eadv8115页

作者： Wang, Yu-Tong Ye, Qi-Hang Yan, Jun-Yong Qiao, Yufei Liu, Yu-Xin Ye, Yong-Zheng Chen, Chen Cheng, Xiao-Tian Li, Chen-Hui Zhang, Zi-Jian Huang, Cheng-Nian Meng, Yun Zou, Kai Zhan, Wen-Kang Zhao, Chao Hu, Xiaolong Tee, Clarence Augustine T.H. Sha, Wei E.I. Huang, Zhixiang Liu, Huiyun Jin, Chao-Yuan Ying, Lei Liu, Feng State Key Laboratory of Extreme Photonics and Instrumentation College of Information Science and Electronic Engineering Zhejiang University Hangzhou310027 China School of Physics Zhejiang Key Laboratory of Micro-nano Quantum Chips and Quantum Control Zhejiang University Hangzhou310027 China International Joint Innovation Center Zhejiang University Haining314400 China School of Precision Instrument and Optoelectronic Engineering Tianjin University Tianjin300072 China Key Laboratory of Optoelectronic Information Science and Technology Ministry of Education Tianjin300072 China Laboratory of Solid State Optoelectronics Information Technology Institute of Semiconductors Chinese Academy of Sciences Beijing100083 China College of Materials Science and Opto-Electronic Technology University of Chinese Academy of Science Beijing101804 China College of Physics and Electrical Information Engineering Zhejiang Normal University Hangzhou310058 China Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University Hefei230039 China Department of Electronic and Electrical Engineering University College London LondonWC1E 7JE United Kingdom ZJU-Hangzhou Global Scientific and Technological Innovation Center Zhejiang University Zhejiang Hangzhou311200 China

Quantum emitters are a key component in photonic quantum technologies. Enhancing single-photon emission by engineering their photonic environment is essential for improving overall efficiency in quantum information processing. However, this enhancement is often limited by the need for ultraprecise emitter placement within conventional photonic cavities. Inspired by the fascinating physics of moiré pattern, we propose a multilayer moiré photonic crystal with a robust isolated flatband. Theoretical analysis reveals that, with nearly infinite photonic density of states, the moiré cavity simultaneously has a high Purcell factor and large tolerance over the emitter’s position, breaking the constraints of conventional cavities. We then experimentally demonstrate various cavity quantum electrodynamic phenomena with a quantum dot in moiré cavity. A large tuning range (up to 40-fold) of quantum dot’s radiative lifetime is achieved through strong Purcell enhancement and inhibition effects. Our findings open the door for moiré flatband cavity–enhanced quantum light sources and quantum nodes for the quantum internet. Copyright © 2025 The Authors, some rights reserved.

关键词： Quantum electronics

STk: a scalable module for solving top-k problems 24

学校读者我要写书评

暂无评论

STk: a scalable module for solving top-k problems

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Hanchen Xia Weidong Liu Xiaojun Mao School of Mathematical Sciences Shanghai Jiao Tong University Shanghai China and RoyalFlush AI Research Institute Hangzhou China School of Mathematical Sciences Shanghai Jiao Tong University Shanghai China and Ministry of Education Key Lab of Artificial Intelligence Shanghai Jiao Tong University Shanghai China School of Mathematical Sciences Shanghai Jiao Tong University Shanghai China and Ministry of Education Key Laboratory of Scientific and Engineering Computing Shanghai Jiao Tong University Shanghai China

ISBN: (纸本)9798331314385

The cost of ranking becomes significant in the new stage of deep learning. We propose STk, a fully differentiable module with a single trainable parameter, designed to solve the Top-k problem without requiring additional time or GPU memory. Due to its fully differentiable nature, STk can be embedded end-to-end into neural networks and optimize the Top-k problems within a unified computational graph. We apply STk to the Average Top-k Loss (ATk), which inherently faces a Top-k problem. The proposed STk Loss outperforms ATk Loss and achieves the best average performance on multiple benchmarks, with the lowest standard deviation. With the assistance of STk Loss, we surpass the state-of-the-art (SOTA) on both CIFAR-100-LT and Places-LT leaderboards.

关键词：

Subspace method based on neural networks for eigenvalue problems

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Dai, Xiaoying Fan, Yunying Sheng, Zhiqiang LSEC Institute of Computational Mathematics and Scientific/Engineering Computing Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing100190 China School of Mathematical Sciences University of Chinese Academy of Sciences Beijing100049 China National Key Laboratory of Computational Physics Institute of Applied Physics and Computational Mathematics Beijing100088 China HEDPS Center for Applied Physics and Technology College of Engineering Peking University Beijing100871 China

With the rapid development of machine learning, numerical discretization methods based on deep neural networks have been widely used in many fields, especially in solving high-dimensional problems where traditional methods face bottlenecks. However, for low-dimensional problems, existing machine learning methods are not as accurate and efficient as traditional methods. In this paper, we propose a subspace method based on neural networks for eigenvalue problems with high accuracy and low cost. Our basic idea is to use neural network based basis functions to span a subspace, then calculate the parameters of the neural network based basis functions through appropriate training, and finally calculate the Galerkin projection of the eigenvalue problem onto the subspace and obtain an approximate solution. In addition, we reduce the dimension of the subspace by applying some dimensionality reduction technique, which can improve the accuracy of the approximate solution in further. Numerical experiments show that we can obtain approximate eigenvalues with accuracy of 10−11 but with less then 400 epochs, which is significantly superior to other existing neural network based *** Codes 65N25, 65L15, 68T07 Copyright © 2024, The Authors. All rights reserved.

关键词：