On Convergence of Gradient Expected Sarsa(λ)

Authors

  • Long Yang Zhejiang University, China
  • Gang Zheng Zhejiang University, China
  • Yu Zhang Zhejiang University, China
  • Qian Zheng Nanyang Technological University,Singapore
  • Pengfei Li Zhejiang University, China
  • Gang Pan Zhejiang University, China

Keywords:

Reinforcement Learning

Abstract

We study the convergence of Expected Sarsa(λ) with function approximation. We show that with off-line es- timate (multi-step bootstrapping) to ExpectedSarsa(λ) is unstable for off-policy learning. Furthermore, based on convex-concave saddle-point framework, we propose a con- vergent Gradient Expected Sarsa(λ) (GES(λ)) algorithm. The theoretical analysis shows that the proposed GES(λ) converges to the optimal solution at a linear convergence rate under true gradient setting. Furthermore, we develop a Lyapunov function technique to investigate how the step- size influences finite-time performance of GES(λ). Addition- ally, such a technique of Lyapunov function can be poten- tially generalized to other gradient temporal difference algo- rithms. Finally, our experiments verify the effectiveness of our GES(λ). For the details of proof, please refer to https: //arxiv.org/pdf/2012.07199.pdf.

Downloads

Published

2021-05-18

How to Cite

Yang, L., Zheng, G., Zhang, Y., Zheng, Q., Li, P., & Pan, G. (2021). On Convergence of Gradient Expected Sarsa(λ). Proceedings of the AAAI Conference on Artificial Intelligence, 35(12), 10621-10629. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/17270

Issue

Section

AAAI Technical Track on Machine Learning V