典型文献
Plume:Lightweight and Generalized Congestion Control with Deep Reinforcement Learning
文献摘要：
Congestion control(CC)is always an important issue in the field of networking,and the enthusiasm for its research has never diminished in both academia and industry.In current years,due to the rapid development of machine learning(ML),the combination of reinforcement learning(RL)and CC has a striking effect.However,These complicated schemes lack generalization and are too heavyweight in storage and computing to be directly implemented in mobile devices.In order to address these problems,we propose Plume,a high-performance,lightweight and generalized RL-CC scheme.Plume proposes a lightweight framework to reduce the overheads while preserving the original performance.Besides,Plume innovatively modifies the framework parameters of the reward function during the retraining process,so that the algorithm can be applied to a variety of scenarios.Evaluation results show that Plume can retain almost all the performance of the original model but the size and decision latency can be reduced by more than 50％and 20％,respectively.Moreover,Plume has better performances in some special scenes.
文献关键词：
中图分类号：
作者姓名：
Dehui Wei;Jiao Zhang;Xuan Zhang;Chengyuan Huang
作者机构：
State Key Laboratory of Networking and Switching Technology,BUPT 100876,China
文献出处：
引用格式：
[1]Dehui Wei;Jiao Zhang;Xuan Zhang;Chengyuan Huang-.Plume:Lightweight and Generalized Congestion Control with Deep Reinforcement Learning)[J].中国通信（英文版）,2022(12):101-117
A类：
Plume
B类：
Lightweight,Generalized,Congestion,Control,Deep,Reinforcement,Learning,control,CC,always,important,issue,field,networking,enthusiasm,its,research,has,never,diminished,both,academia,industry,In,current,years,due,rapid,development,machine,learning,ML,combination,reinforcement,RL,striking,effect,However,These,complicated,schemes,lack,generalization,are,too,heavyweight,storage,computing,directly,implemented,mobile,devices,order,address,these,problems,high,lightweight,generalized,proposes,framework,overheads,while,preserving,original,Besides,innovatively,modifies,parameters,reward,function,during,retraining,process,that,algorithm,can,applied,variety,scenarios,Evaluation,results,show,retain,almost,all,model,but,size,decision,latency,reduced,by,more,than,respectively,Moreover,better,performances,some,special,scenes
AB值：
0.657701
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
