检索结果-内蒙古大学图书馆

Knowledge-Guided Semantic Transfer Network for Few-Shot Image Recognition

ieee TRANSACTIONS ON NEURAL NETWORKS AND learning SYSTEMS 2023年第2期PP卷 3573-3586页

作者： Li, Zechao Tang, Hao Peng, Zhimao Qi, Guo-Jun Tang, Jinhui Nanjing Univ Sci & Technol Sch Comp Sci & Engn Nanjing 210094 Peoples R China Nankai Univ Coll Comp Sci Tianjin 300350 Peoples R China Maple Lab Net Lab MAchine Percept & Learning MAPLE Orlando FL 32816 USA Futurewei Technol Seattle WA 98004 USA

Deep learning-based models have been shown to outperform human beings in many computer vision tasks with massive available labeled training data in learning. However, humans have an amazing ability to easily recognize images of novel categories by browsing only a few examples of these categories. In this case, few-shot learning comes into being to make machines learn from extremely limited labeled examples. One possible reason why human beings can well learn novel concepts quickly and efficiently is that they have sufficient visual and semantic prior knowledge. Toward this end, this work proposes a novel knowledge-guided semantic transfer network (KSTNet) for few-shot image recognition from a supplementary perspective by introducing auxiliary prior knowledge. The proposed network jointly incorporates vision inferring, knowledge transferring, and classifier learning into one unified framework for optimal compatibility. A category-guided visual learning module is developed in which a visual classifier is learned based on the feature extractor along with the cosine similarity and contrastive loss optimization. To fully explore prior knowledge of category correlations, a knowledge transfer network is then developed to propagate knowledge information among all categories to learn the semantic-visual mapping, thus inferring a knowledge-based classifier for novel categories from base categories. Finally, we design an adaptive fusion scheme to infer the desired classifiers by effectively integrating the above knowledge and visual information. Extensive experiments are conducted on two widely used Mini-ImageNet and Tiered-ImageNet benchmarks to validate the effectiveness of KSTNet. Compared with the state of the art, the results show that the proposed method achieves favorable performance with minimal bells and whistles, especially in the case of one-shot learning.

关键词： Visualization Semantics Image recognition Task analysis Knowledge transfer Knowledge engineering Feature extraction Few-shot learning image recognition knowledge graph knowledge transfer

来源：评论

学校读者我要写书评

暂无评论

Receding Horizon Actor-Critic learning Control for Nonlinear Time-Delay Systems With Unknown dynamics

引用

ieee TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2023年第8期53卷 4980-4993页

作者： Liu, Jiahang Zhang, Xinglong Xu, Xin Xiong, Quan Natl Univ Def Technol Coll Intelligence Sci & Technol Changsha 410073 Peoples R China Beijing Inst Biotechnol Beijing 100071 Peoples R China

With the development of modern mechatronics and networked systems, the controller design of time-delay systems has received notable attention. Time delays can greatly influence the stability and performance of the systems, especially for optimal control design. In this article, we propose a receding horizon actor-critic learning control approach for near-optimal control of nonlinear time-delay systems (RACL-TD) with unknown dynamics. In the proposed approach, a data-driven predictor for nonlinear time-delay systems is first learned based on the Koopman theory using precollected samples. Then, a receding horizon actor-critic architecture is designed to learn a near-optimal control policy. In RACL-TD, the terminal cost is determined by using the Lyapunov-Krasovskii approach so that the influences of the delayed states and control inputs can be well addressed. Furthermore, a relaxed terminal condition is present to reduce the computational cost. The convergence and optimality of RACL-TD in each prediction interval as well as the closed-loop property of the system are discussed and analyzed. Simulation results on a two-stage time-delayed chemical reactor illustrate that RACL-TD can achieve better control performance than nonlinear model predictive control (MPC) and infinite-horizon adaptive dynamic programming. Moreover, RACL-TD can have less computational cost than nonlinear MPC.

关键词： Delay effects Optimal control Control systems Stability criteria Simulation Predictive control Costs Discrete-time nonlinear systems Koopman operator receding horizon control reinforcement learning (RL) time-delay systems

来源：评论

学校读者我要写书评

暂无评论

adaptive Optimal Output Regulation of Discrete-Time Linear Systems: A reinforcement learning Approach 62

Adaptive Optimal Output Regulation of Discrete-Time Linear S...

引用

62nd ieee Conference on Decision and Control (CDC)

作者： Chakraborty, Sayan Gao, Weinan Vamvoudakis, Kyriakos G. Jiang, Zhong-Ping NYU CAN Lab Brooklyn NY 11201 USA Northeastern Univ State Key Lab Synthet Automat Proc Ind Shenyang 110819 Peoples R China Georgia Inst Technol Daniel Guggenheim Sch Aerosp Engn Atlanta GA 30332 USA

ISBN: (纸本)9798350301243

In this paper, we solve the optimal output regulation problem for discrete-time systems without precise knowledge of the system model. Drawing inspiration from reinforcement learning and adaptive dynamic programming, a data-driven solution is developed that enables asymptotic tracking and disturbance rejection. Notably, it is discovered that the proposed approach for discrete-time output regulation differs from the continuous-time approach in terms of the persistent excitation condition required for policy iteration to be unique and convergent. To address this issue, a new persistent excitation condition is introduced to ensure both uniqueness and convergence of the data-driven policy iteration. The efficacy of the proposed methodology is validated by an inverted pendulum on a cart example.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Energy Management Strategy for Fuel Cell/Battery/Ultracapacitor Hybrid Electric Vehicles Using Deep reinforcement learning With Action Trimming

引用

ieee TRANSACTIONS ON VEHICULAR TECHNOLOGY 2022年第7期71卷 7171-7185页

作者： Fu, Zhumu Wang, Haocong Tao, Fazhan Ji, Baofeng Dong, Yongsheng Song, Shuzhong Henan Univ Sci & Technol Coll Informat Engn Luoyang 471023 Peoples R China Henan Univ Sci & Technol Henan Key Lab Robot & Intelligent Syst Luoyang 471023 Peoples R China

As for fuel cell hybrid electric vehicle equipped with battery (BAT) and ultracapacitor (UC), its dynamic topology structure is complex and different characteristics of three power sources induce challenges in energy management for fuel economy, power sources lifespan, and dynamic performance of the vehicle. In this paper, an energy management strategy (EMS) based on a hierarchical power splitting structure and deep reinforcement learning (DRL) is proposed. In the higher layer strategy of the proposed EMS, the UC is employed to supply peak power and recover braking energy through the adaptive filter based on fuzzy control. Then, the integrated DRL and equivalent consumption minimization strategy framework is proposed to optimize the power allocation of fuel cell (FC) and BAT in the lower layer, to ensure the highly efficient operation of FC and reduce hydrogen consumption. And the action trimming based on heuristic technique is proposed to further restrain the adverse effect of sudden peak power on FC lifespan. The simulation results show the proposed EMS can make the output of FC smoother, improve its working efficiency to alleviate the stress of BAT, and increase by 14.8% compared with the Q-learning strategy in fuel economy under WLTP driving cycle. Meanwhile, the obtained results under UDDSHDV show fuel economy of the proposed EMS can reach dynamic programming (DP) benchmark level of 89.7%.

关键词： Dater driven deep reinforcement learning energy management strategy fuel cell hybrid electric vehicle heuristic technique

来源：评论

学校读者我要写书评

暂无评论

Event-Triggered Optimal Tracking Control for Uncertain Nonlinear System Based on reinforcement learning 14

Event-Triggered Optimal Tracking Control for Uncertain Nonli...

引用

14th International Conference on Information Science and Technology, ICIST 2024

作者： Wang, Yuanhao Bai, Weiwei Navigation College Dalian Maritime University Dalian China

ISBN: (纸本)9798350353334

In this paper, an event-triggered optimal tracking control problem is studied for uncertain nonlinear system based on reinforcement learning (RL). Firstly, a class of nonlinear dynamic system with general uncertainty is considered and the augmented system comprising tracking error and reference signal is constructed. Secondly, an improved adaptive dynamic programming (ADP) technique, involving actor-critic algorithm and fuzzy logic systems, is developed to solve the Hamilton-Jacobi-Bellman (HJB) equation with respect to nominal augmented system. Thirdly, in order to reduce the mechanical wear of actuator and energy consumption, event-triggered mechanism is performed in controller updating. Finally, stability analysis proofs that all signals are uniformly ultimately bounded (UUB) in the closed-loop system via Lyapunov theory. Simulation results verify feasibility of proposed scheme. © 2024 ieee.

关键词： reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Intelligent Energy Management Strategy Based on an Improved reinforcement learning Algorithm With Exploration Factor for a Plug-in PHEV

引用

ieee TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2022年第7期23卷 8725-8735页

作者： Lin, Xinyou Zhou, Kuncheng Mo, Liping Li, Hailin Fuzhou Univ Coll Mech Engn & Automat Fuzhou 350108 Peoples R China West Virginia Univ Dept Mech & Aerosp Engn Morgantown WV 26506 USA

An intelligent energy management strategy (EMS) based on an improved reinforcement learning (RL) algorithm is developed to enhance the adaptability of the EMS and to further improve the fuel efficiency of a Plug-in Parallel Hybrid Electric Vehicle (PHEV). Both the numerical model and the energy management strategy of a plug-in PHEV are described. The improved RL with Q-learning algorithm is implemented to acquire the optimal control strategies for improving fuel economy. The Markov Chain is employed to calculate the Transition Probability Matrix of the required power. A Kullback-Leibler (KL) divergence rate is designed to activate the update of EMS, when a new corresponding driving cycle is expected. An Exploration Factor (FT) is proposed to overcome the disadvantages of the normal RL algorithm in convergence rate and reward cost evaluation. The diverse KL divergence rates are examined to seek optimal solutions. The normal-RI, strategy, rule-based strategy, and dynamic programming strategy are implemented as benchmark strategies to verify the effectiveness of the proposed strategy. The validation results indicate that the improved RL algorithm with FY makes it possible to promote the EMS capable of significantly improving the energy efficiency of a plug-in PHEV.

关键词： Hybrid electric vehicle energy management strategy reinforcement learning algorithm self-adaptive exploration factor

来源：评论

学校读者我要写书评

暂无评论

Improve Video Conferencing Quality with Deep reinforcement learning 36

Improve Video Conferencing Quality with Deep Reinforcement L...

引用

36th ieee/IFIP Network Operations and Management symposium, NOMS 2023

作者： Tu, Nguyen Van Ko, Kyungchan Ryu, Sangwoo Ha, Sangtae Hong, James Won-Ki Postech Dept. of Computer Science and Engineering Pohang Korea Republic of Postech Graduate School of Artificial Intelligence Pohang Korea Republic of University of Colorado Boulder Dept. of Computer Science CO United States

ISBN: (纸本)9781665477161

Many studies have applied machine learning to bitrate control to increase Quality of Experience (QoE) of video streaming services in highly dynamic networks. However, their solutions mainly focused on HTTP adaptive streaming with one-to-one connections. This paper studies video conferencing applications where multi-party, full-duplex communication happens among participants. In particular, we propose Muno, a Deep reinforcement learning (DRL)-based bandwidth prediction framework for multi-party video conferencing systems. Muno learns and predicts an appropriate bitrate for each connection in a multi-party conferencing call. We trained Muno to maximize the QoE of individual connections by constructing a feedback loop between a media server and DRL servers. Our experimental results show that Muno achieves a higher video streaming rate and lower delay compared to state-of-the-art rulebased algorithms. © 2023 ieee.

关键词： reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Behavior Replication of Cascaded dynamic Systems Using Machine learning

Behavior Replication of Cascaded Dynamic Systems Using Machi...

引用

2023 ieee International symposium on Robotic and Sensors Environments, ROSE 2023

作者： Abouheaf, Mohammed Gueaieb, Wail Mayyas, Mohammad Bowling Green State University Robotics Engineering Bowling GreenOH43403 United States Ottawa University School of Electrical Engineering and Computer Science OttawaONK1N 6N5 Canada

ISBN: (纸本)9798350308044

This paper introduces an innovative data-driven approach for replicating behaviors in interconnected and heterogeneous dynamic systems. The core concept involves real-time control of dynamic systems to closely mimic reference-model trajectories using model-free techniques. Within this coupled framework, one component possesses complete information about reference-trajectories, although not necessarily their dynamics. In contrast, follower systems, with limited connectivity to reference-model trajectories, exclusively replicate the behavior of the primary process, which retains insight into model-reference dynamics. The adopted strategies are causal, integrating higher-order error dynamics to ensure precise tracking of reference-trajectories. Furthermore, these strategies incorporate variations in reference-model dynamics via a pseudo partial derivative, akin to sensitivity derivatives in model-reference adaptive strategies. To optimize the dynamic behavior of the follower process, the solution employs a reinforcement learning mechanism through adaptive critics. This mechanism approximates the optimal strategy and the associated value function. The actor and critic weights of the adaptive critic structure are tuned using a projection technique to ensure convergence of the adapted strategy. The validation of this solution is demonstrated on a dynamic system with delays, simulating an underwater vehicle scenario. The developed methodology is rigorously compared with another high-order model-free adaptive control approach. The presented approach showcases its capability to effectively replicate behaviors, resulting in improved tracking accuracy. © 2023 ieee.

关键词： Trajectories

来源：评论

学校读者我要写书评

暂无评论

Online Service Provisioning in NFV-Enabled Networks Using Deep reinforcement learning

引用

ieee TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT 2022年第3期19卷 3276-3289页

作者： Nouruzi, Ali Zakeri, Abolfazl Javan, Mohammad Reza Mokari, Nader Hussain, Rasheed Kazmi, S. M. Ahsan Tarbiat Modares Univ Dept Elect & Comp Engn Tehran 14115111 Iran Shahrood Univ Technol Dept Elect Engn Shahrood *** Iran Univ Bristol Smart Internet Lab Bristol BS8 1TH Avon England Univ Bristol Bristol Digital Futures Inst Bristol BS8 1TH Avon England Univ West England Fac Comp Sci & Creat Technol Bristol BS16 1QY Avon England

In this paper, we study a Deep reinforcement learning (DRL) based framework for an online end-user service provisioning in a Network Function Virtualization (NFV)-enabled network. We formulate an optimization problem aiming to minimize the cost of network resource utilization. The main challenge is provisioning the online service requests by fulfilling their Quality of Service (QoS) under limited resource availability. Moreover, fulfilling the stochastic service requests in a large network is another challenge that is evaluated in this paper. To solve the formulated optimization problem in an efficient and intelligent manner, we propose a Deep Q-Network for adaptive Resource allocation (DQN-AR) in NFV-enabled network for function placement and dynamic routing which considers the available network resources as DQN states. Moreover, the service's characteristics, including the service life time and number of the arrival requests, are modeled by the Uniform and Exponential distribution, respectively. In addition, we evaluate the computational complexity of the proposed method. Numerical results carried out for different ranges of parameters reveal the effectiveness of our framework. In specific, the obtained results show that the average number of admitted requests of the network increases by 7 up to 14% and the network utilization cost decreases by 5 and 20%.

关键词： Routing Heuristic algorithms dynamic scheduling Resource management Quality of service dynamic programming Delays Deep reinforcement learning service lifetime resource allocation NFV

来源：评论

学校读者我要写书评

暂无评论

dynamic Resource Management for Cloud-native Bulk Synchronous Parallel Applications 26

Dynamic Resource Management for Cloud-native Bulk Synchronou...

引用

26th ieee International symposium on Real-Time Distributed Computing (ISORC)

作者： Wang, Evan Barve, Yogesh Gokhale, Aniruddha Sun, Hongyang Vanderbilt Univ Dept CS 221 Kirkland Hall Nashville TN 37235 USA Univ Kansas Dept EECS Lawrence KS 66045 USA

ISBN: (纸本)9798350339024

Many traditional high-performance computing applications including those that follow the Bulk Synchronous Parallel (BSP) communication paradigm are increasingly being deployed in cloud-native virtualized and multi-tenant container clusters. However, such a shared, virtualized platform limits the degree of control that BSP applications can have in effectively allocating resources. This can adversely impact their performance, particularly when stragglers manifest in individual BSP supersteps. Existing BSP resource management solutions assume the same execution time for individual tasks at every superstep, which is not always the case. To address these limitations, we present a dynamic resource management middleware for cloud-native BSP applications comprising a heuristics algorithm that determines effective resource configurations across multiple supersteps while considering dynamic workloads per superstep, and trading off performance improvements with reconfiguration costs. Moreover, we design dynamic programming and reinforcement learning approaches that can be used as pluggable strategies to determine whether and when to enforce a reconfiguration. Empirical evaluations of our solution show between 10% and 25% improvement in performance over a baseline static approach even in the presence of reconfiguration penalty.

关键词： Bulk Synchronous Parallel jobs Resource management Cloud-native workload forecasting

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：