NSTL国家科技图书文献中心

1. SCORE: Simple Contrastive Representation and Reset-Ensemble for offline meta-reinforcement learning NSTL国家科技图书文献中心

Yang H. | Lin K.... - 《Knowledge-based systems》 - 2025,309(Jan.30) - 1.1～1.10 - 共10页

摘要：© 2024Offline meta-reinforcement learning | learning framework without negative samples to pre-train | during meta-training. Furthermore, SCORE employs a | continual learning* ability and enhance their perception of* | (OMRL) aims to train agents to quickly adapt to new

关键词： Contrastive learning | Offline meta reinforcement learning | Reset-Ensemble

2. Harnessing Meta-Reinforcement Learning for Enhanced Tracking in Geofencing Systems NSTL国家科技图书文献中心

Alireza Famili | Shihua Sun ... - 《IEEE Open Journal of the Communications Society》 - 2025,6 - 944～960 - 共17页

摘要： MetaFence: Meta-Reinforcement* Learning for Geofencing* | are optimally deployed using a meta-reinforcement | learning (meta-RL) framework. Our proposed meta-RL method | -induced errors. Moreover, the meta-training approach | traditional methods. Furthermore, we show that the meta

关键词： 5G mobile communication | Accuracy | Three-dimensional displays | Geometry | Drones | Distance measurement | Wireless fidelity | NP-hard problem | Mixed reality | Metaverse

3. Contrastive meta-reinforcement learning for heterogeneous graph neural architecture search NSTL国家科技图书文献中心

Zixuan Xu | Jia Wu - 《Expert Systems with Application》 - 2025,260(Jan.) - 125433.1～125433.13 - 共13页

摘要： called Contrastive Meta-reinforcement* learning-based* | gradient-based meta-learning* to rapidly adapt to new* | tasks by leveraging the knowledge acquired from meta | -training tasks. Secondly, since our meta-training tasks | learning to unify the evaluation metric across all tasks

关键词： Heterogeneous graph neural architecture search | Meta-learning | Contrastive learning

4. Hierarchical Multi-Agent Meta-Reinforcement Learning for Cross-Channel Bidding NSTL国家科技图书文献中心

Shenghong He | Chao Yu ... - 《IEEE Transactions on Knowledge and Data Engineering》 - 2025,37(3) - 1241～1254 - 共14页

摘要： multi-agent reinforcement* learning framework for multi* | -based meta-channel knowledge learning* method to* | problem of extrapolation errors in offline learning | Real-time bidding (RTB) plays a pivotal role | in online advertising ecosystems. Advertisers

关键词： Resource management | Advertising | Optimization | Hidden Markov models | Reinforcement learning | Real-time systems | Dynamic scheduling | Extrapolation | Training | Search problems

5. Adaptive Jamming Decision-Making Against FHSS Communications via Inexpert Demonstrations Assisted Meta Reinforcement Learning NSTL国家科技图书文献中心

Ning Rao | Hua Xu ... - 《IEEE communications letters》 - 2025,29(1) - 105～109 - 共5页

摘要：Reinforcement learning (RL)s powerful | , we propose a meta* RL method for frequency-hopping* | network is meta-trained with multiple diverse tasks to | demonstrations, along with learning* rate adaptation to achieve* | optimization capabilities have been extensively applied in

关键词： Jamming | Decision making | Bandwidth | Optimization | Reinforcement learning | Cloning | Vectors | Space exploration | Long short term memory | Interference

6. Bionic cooperative load frequency control in interconnected grids: A multi-agent deep Meta reinforcement learning approach NSTL国家科技图书文献中心

Li J. | Dai J.... - 《Applied energy》 - 2025,379(Feb.1) - 1.1～1.13 - 共13页

摘要： Automatic Curriculum Multi-Agent Deep Meta* Actor-Critic* | employs a hybrid curriculum learning* strategy, enabling* | gradual learning* and adaptation, which enhances the* | © 2024In the interconnected power grid | operating within a performance-based frequency regulation

关键词： Automatic curriculum multi-agent deep meta actor critic | Distributed neural network | Interconnected grid | Load frequency control | Regulation mileage payment

7. Efficient Replay Deep Meta-Reinforcement Learning for Active Fault-Tolerant Control of Solid Oxide Fuel Cell Systems Considering Multivariable Coordination NSTL国家科技图书文献中心

Jiawen Li | Tao Zhou - 《IEEE transactions on transportation electrification》 - 2025,11(1) - 4803～4817 - 共15页

摘要： and meta-learning* techniques to improve the* | deep meta-deterministic policy gradient (ER-DMDPG | robustness and multitask cooperative learning* capability of* | , which is trained by a cooperative meta-learner and a | A data-driven integrated active fault-tolerant

关键词： Fault tolerant systems | Fault tolerance | Hydrogen | Voltage control | Fuels | Fuel cells | Temperature control | Prediction algorithms | Heuristic algorithms | Genetic algorithms

8. A meta-heuristic algorithm combined with deep reinforcement learning for multi-sensor positioning layout problem in complex environment NSTL国家科技图书文献中心

Yida Ning | Zhenzu Bai ... - 《Expert Systems with Application》 - 2025,261(Feb.) - 125555.1～125555.16 - 共16页

摘要：In a multi-sensor positioning system (MSPS | ), the layout of sensors plays a crucial role in | determining the system's performance. Therefore, addressing | the sensor layout problem (SLP) within the MSPS is | an essential approach to achieve high-precision

关键词： Multi-sensor positioning system | Sensor layout problem | Constrained multi-objective evolutionary algorithm | Deep Q network | Multi-operator reproduction

9. Meta-learning-based fault-tolerant attitude control of hypersonic flight vehicle with input constraints NSTL国家科技图书文献中心

Xiaoxiang,Hu | Kejun,Dong ... - 《Nonlinear dynamics》 - 2025,113(1) - 711～728 - 共18页

摘要：Abstract In this study, a meta-learning-based | advantage of integral reinforcement* learning (IRL* | ) algorithm and meta-learning, can greatly reduce the | control law. The meta-learning* ideas are also adopted to* | improved by meta-learning. The ultimately uniformly

关键词： Hypersonic flight vehicle | Attitude tracking control | Fault-tolerant control | Integral reinforcement learning | Meta-learning

10. A Survey on Deep Learning-Based Traffic Signal Control NSTL国家科技图书文献中心

Qinbatu Si | Lirun Yang ... - 《Journal of circuits, systems and computers》 - 2025,34(2) - 1.1～1.45 - 共45页

摘要： Reinforcement Learning, Federated Learning, and Meta-learning | Reinforcement Learning in this field, there remains a notable | researchers are turning to Deep Learning* (DL) methods to* | Intelligent Traffic Management is a crucial | issue closely related to daily life and productivity

摘要：© 2024Offline meta-reinforcement learning | learning framework without negative samples to pre-train | during meta-training. Furthermore, SCORE employs a | continual learning* ability and enhance their perception of* | (OMRL) aims to train agents to quickly adapt to new

关键词： Contrastive learning | Offline meta reinforcement learning | Reset-Ensemble

摘要： MetaFence: Meta-Reinforcement* Learning for Geofencing* | are optimally deployed using a meta-reinforcement | learning (meta-RL) framework. Our proposed meta-RL method | -induced errors. Moreover, the meta-training approach | traditional methods. Furthermore, we show that the meta

关键词： 5G mobile communication | Accuracy | Three-dimensional displays | Geometry | Drones | Distance measurement | Wireless fidelity | NP-hard problem | Mixed reality | Metaverse

摘要： called Contrastive Meta-reinforcement* learning-based* | gradient-based meta-learning* to rapidly adapt to new* | tasks by leveraging the knowledge acquired from meta | -training tasks. Secondly, since our meta-training tasks | learning to unify the evaluation metric across all tasks

关键词： Heterogeneous graph neural architecture search | Meta-learning | Contrastive learning

摘要： multi-agent reinforcement* learning framework for multi* | -based meta-channel knowledge learning* method to* | problem of extrapolation errors in offline learning | Real-time bidding (RTB) plays a pivotal role | in online advertising ecosystems. Advertisers

关键词： Resource management | Advertising | Optimization | Hidden Markov models | Reinforcement learning | Real-time systems | Dynamic scheduling | Extrapolation | Training | Search problems

摘要：Reinforcement learning (RL)s powerful | , we propose a meta* RL method for frequency-hopping* | network is meta-trained with multiple diverse tasks to | demonstrations, along with learning* rate adaptation to achieve* | optimization capabilities have been extensively applied in

关键词： Jamming | Decision making | Bandwidth | Optimization | Reinforcement learning | Cloning | Vectors | Space exploration | Long short term memory | Interference

摘要： Automatic Curriculum Multi-Agent Deep Meta* Actor-Critic* | employs a hybrid curriculum learning* strategy, enabling* | gradual learning* and adaptation, which enhances the* | © 2024In the interconnected power grid | operating within a performance-based frequency regulation

关键词： Automatic curriculum multi-agent deep meta actor critic | Distributed neural network | Interconnected grid | Load frequency control | Regulation mileage payment

摘要： and meta-learning* techniques to improve the* | deep meta-deterministic policy gradient (ER-DMDPG | robustness and multitask cooperative learning* capability of* | , which is trained by a cooperative meta-learner and a | A data-driven integrated active fault-tolerant

关键词： Fault tolerant systems | Fault tolerance | Hydrogen | Voltage control | Fuels | Fuel cells | Temperature control | Prediction algorithms | Heuristic algorithms | Genetic algorithms

摘要：In a multi-sensor positioning system (MSPS | ), the layout of sensors plays a crucial role in | determining the system's performance. Therefore, addressing | the sensor layout problem (SLP) within the MSPS is | an essential approach to achieve high-precision

关键词： Multi-sensor positioning system | Sensor layout problem | Constrained multi-objective evolutionary algorithm | Deep Q network | Multi-operator reproduction

摘要：Abstract In this study, a meta-learning-based | advantage of integral reinforcement* learning (IRL* | ) algorithm and meta-learning, can greatly reduce the | control law. The meta-learning* ideas are also adopted to* | improved by meta-learning. The ultimately uniformly

关键词： Hypersonic flight vehicle | Attitude tracking control | Fault-tolerant control | Integral reinforcement learning | Meta-learning

摘要： Reinforcement Learning, Federated Learning, and Meta-learning | Reinforcement Learning in this field, there remains a notable | researchers are turning to Deep Learning* (DL) methods to* | Intelligent Traffic Management is a crucial | issue closely related to daily life and productivity

关键词： Traffic Signal Control | intelligent transportation system | Deep Reinforcement Learning | Federated Learning | Meta-learning

4008-161-200 800-990-8900

国家科技图书文献中心

摘要：© 2024Offline meta-reinforcement learning | learning framework without negative samples to pre-train | during meta-training. Furthermore, SCORE employs a | continual learning ability and enhance their perception of | (OMRL) aims to train agents to quickly adapt to new

关键词： Contrastive learning | Offline meta reinforcement learning | Reset-Ensemble

摘要： MetaFence: Meta-Reinforcement Learning for Geofencing | are optimally deployed using a meta-reinforcement | learning (meta-RL) framework. Our proposed meta-RL method | -induced errors. Moreover, the meta-training approach | traditional methods. Furthermore, we show that the meta

关键词： 5G mobile communication | Accuracy | Three-dimensional displays | Geometry | Drones | Distance measurement | Wireless fidelity | NP-hard problem | Mixed reality | Metaverse

摘要： called Contrastive Meta-reinforcement learning-based | gradient-based meta-learning to rapidly adapt to new | tasks by leveraging the knowledge acquired from meta | -training tasks. Secondly, since our meta-training tasks | learning to unify the evaluation metric across all tasks

关键词： Heterogeneous graph neural architecture search | Meta-learning | Contrastive learning

摘要： multi-agent reinforcement learning framework for multi | -based meta-channel knowledge learning method to | problem of extrapolation errors in offline learning | Real-time bidding (RTB) plays a pivotal role | in online advertising ecosystems. Advertisers

关键词： Resource management | Advertising | Optimization | Hidden Markov models | Reinforcement learning | Real-time systems | Dynamic scheduling | Extrapolation | Training | Search problems

摘要：Reinforcement learning (RL)s powerful | , we propose a meta RL method for frequency-hopping | network is meta-trained with multiple diverse tasks to | demonstrations, along with learning rate adaptation to achieve | optimization capabilities have been extensively applied in

关键词： Jamming | Decision making | Bandwidth | Optimization | Reinforcement learning | Cloning | Vectors | Space exploration | Long short term memory | Interference

摘要： Automatic Curriculum Multi-Agent Deep Meta Actor-Critic | employs a hybrid curriculum learning strategy, enabling | gradual learning and adaptation, which enhances the | © 2024In the interconnected power grid | operating within a performance-based frequency regulation

关键词： Automatic curriculum multi-agent deep meta actor critic | Distributed neural network | Interconnected grid | Load frequency control | Regulation mileage payment

摘要： and meta-learning techniques to improve the | deep meta-deterministic policy gradient (ER-DMDPG | robustness and multitask cooperative learning capability of | , which is trained by a cooperative meta-learner and a | A data-driven integrated active fault-tolerant

关键词： Fault tolerant systems | Fault tolerance | Hydrogen | Voltage control | Fuels | Fuel cells | Temperature control | Prediction algorithms | Heuristic algorithms | Genetic algorithms

摘要：In a multi-sensor positioning system (MSPS | ), the layout of sensors plays a crucial role in | determining the system's performance. Therefore, addressing | the sensor layout problem (SLP) within the MSPS is | an essential approach to achieve high-precision

关键词： Multi-sensor positioning system | Sensor layout problem | Constrained multi-objective evolutionary algorithm | Deep Q network | Multi-operator reproduction

摘要：Abstract In this study, a meta-learning-based | advantage of integral reinforcement learning (IRL | ) algorithm and meta-learning, can greatly reduce the | control law. The meta-learning ideas are also adopted to | improved by meta-learning. The ultimately uniformly

关键词： Hypersonic flight vehicle | Attitude tracking control | Fault-tolerant control | Integral reinforcement learning | Meta-learning

摘要： Reinforcement Learning, Federated Learning, and Meta-learning | Reinforcement Learning in this field, there remains a notable | researchers are turning to Deep Learning (DL) methods to | Intelligent Traffic Management is a crucial | issue closely related to daily life and productivity

关键词： Traffic Signal Control | intelligent transportation system | Deep Reinforcement Learning | Federated Learning | Meta-learning

4008-161-200 800-990-8900

国家科技图书文献中心

摘要：© 2024Offline meta-reinforcement learning | learning framework without negative samples to pre-train | during meta-training. Furthermore, SCORE employs a | continual learning* ability and enhance their perception of* | (OMRL) aims to train agents to quickly adapt to new

摘要： MetaFence: Meta-Reinforcement* Learning for Geofencing* | are optimally deployed using a meta-reinforcement | learning (meta-RL) framework. Our proposed meta-RL method | -induced errors. Moreover, the meta-training approach | traditional methods. Furthermore, we show that the meta

摘要： called Contrastive Meta-reinforcement* learning-based* | gradient-based meta-learning* to rapidly adapt to new* | tasks by leveraging the knowledge acquired from meta | -training tasks. Secondly, since our meta-training tasks | learning to unify the evaluation metric across all tasks

摘要： multi-agent reinforcement* learning framework for multi* | -based meta-channel knowledge learning* method to* | problem of extrapolation errors in offline learning | Real-time bidding (RTB) plays a pivotal role | in online advertising ecosystems. Advertisers

摘要：Reinforcement learning (RL)s powerful | , we propose a meta* RL method for frequency-hopping* | network is meta-trained with multiple diverse tasks to | demonstrations, along with learning* rate adaptation to achieve* | optimization capabilities have been extensively applied in

摘要： Automatic Curriculum Multi-Agent Deep Meta* Actor-Critic* | employs a hybrid curriculum learning* strategy, enabling* | gradual learning* and adaptation, which enhances the* | © 2024In the interconnected power grid | operating within a performance-based frequency regulation

摘要： and meta-learning* techniques to improve the* | deep meta-deterministic policy gradient (ER-DMDPG | robustness and multitask cooperative learning* capability of* | , which is trained by a cooperative meta-learner and a | A data-driven integrated active fault-tolerant

摘要：Abstract In this study, a meta-learning-based | advantage of integral reinforcement* learning (IRL* | ) algorithm and meta-learning, can greatly reduce the | control law. The meta-learning* ideas are also adopted to* | improved by meta-learning. The ultimately uniformly

摘要： Reinforcement Learning, Federated Learning, and Meta-learning | Reinforcement Learning in this field, there remains a notable | researchers are turning to Deep Learning* (DL) methods to* | Intelligent Traffic Management is a crucial | issue closely related to daily life and productivity