典型文献
Multi-Agent Few-Shot Meta Reinforcement Learning for Trajectory Design and Channel Selection in UAV-Assisted Networks
文献摘要：
Unmanned aerial vehicle(UAV)-assisted communications have been considered as a solution of aerial networking in future wireless networks due to its low-cost,high-mobility,and swift features.This pa-per considers a UAV-assisted downlink transmission,where UAVs are deployed as aerial base stations to serve ground users.To maximize the average trans-mission rate among the ground users,this paper for-mulates a joint optimization problem of UAV trajec-tory design and channel selection,which is NP-hard and non-convex.To solve the problem,we propose a multi-agent deep Q-network(MADQN)scheme.Specifically,the agents that the UAVs act as per-form actions from their observations distributively and share the same reward.To tackle the tasks where the experience is insufficient,we propose a multi-agent meta reinforcement learning algorithm to fast adapt to the new tasks.By pretraining the tasks with sim-ilar distribution,the learning model can acquire gen-eral knowledge.Simulation results have indicated the MADQN scheme can achieve higher throughput than fixed allocation.Furthermore,our proposed multi-agent meta reinforcement learning algorithm learns the new tasks much faster compared with the MADQN scheme.
文献关键词：
中图分类号：
作者姓名：
Shiyang Zhou;Yufan Cheng;Xia Lei;Huanhuan Duan
作者机构：
National Key Laboratory of Science and Technology on Communications,University of Electronic Science and Technology of China,Chengdu 611731,China
文献出处：
引用格式：
[1]Shiyang Zhou;Yufan Cheng;Xia Lei;Huanhuan Duan-.Multi-Agent Few-Shot Meta Reinforcement Learning for Trajectory Design and Channel Selection in UAV-Assisted Networks)[J].中国通信（英文版）,2022(04):166-176
A类：
mulates,MADQN
B类：
Multi,Agent,Few,Shot,Reinforcement,Learning,Trajectory,Design,Channel,Selection,Assisted,Networks,Unmanned,aerial,vehicle,assisted,communications,have,been,considered,solution,networking,future,wireless,networks,due,its,low,cost,mobility,swift,features,This,considers,downlink,transmission,where,UAVs,deployed,base,stations,serve,ground,users,To,maximize,average,rate,among,this,paper,joint,optimization,problem,trajec,design,channel,selection,which,NP,hard,convex,solve,we,multi,deep,scheme,Specifically,agents,that,form,actions,from,their,observations,distributively,share,same,reward,tackle,tasks,experience,insufficient,meta,reinforcement,learning,algorithm,adapt,new,By,pretraining,sim,ilar,distribution,model,can,acquire,eral,knowledge,Simulation,results,indicated,achieve,higher,throughput,than,fixed,allocation,Furthermore,our,proposed,learns,much,faster,compared
AB值：
0.618733
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。
