Gradient compensation traces based temporal difference learning

Wang Bi,Li Xuelian,Gao Zhiqiang,Chen Yang

Neurocomputing（2021）

引用 2|浏览5

暂无评分

摘要

•An implementation of eligibility traces, named GCT, is proposed.•GCT maintains a long term credit-assignment without the suffering of λ.•The convergence of GCT based TD (named GCTD) is held theoretically.•The data efficiency of GCTD is empirically improved in the sparse reward setting.

查看译文

关键词

Reinforcement learning,Eligibility traces,Value iteration,Temporal difference learning

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要