Paper ID | SPTM-3.5 | ||
Paper Title | Byzantine-Resilient Decentralized TD Learning with Linear Function Approximation | ||
Authors | Zhaoxian Wu, Sun Yat-Sen University, China; Han Shen, Tianyi Chen, Rensselaer Polytechnic Institute, United States; Qing Ling, Sun Yat-Sen University, China | ||
Session | SPTM-3: Estimation, Detection and Learning over Networks 1 | ||
Location | Gather.Town | ||
Session Time: | Tuesday, 08 June, 14:00 - 14:45 | ||
Presentation Time: | Tuesday, 08 June, 14:00 - 14:45 | ||
Presentation | Poster | ||
Topic | Signal Processing Theory and Methods: Signal Processing over Networks | ||
IEEE Xplore Open Preview | Click here to view in IEEE Xplore | ||
Abstract | This paper considers the policy evaluation problem in reinforcement learning with agents of a decentralized and directed network. The focus is on decentralized temporal-difference (TD) learning with linear function approximation in the presence of unreliable or even malicious agents, termed as Byzantine agents. In order to evaluate the quality of a fixed policy in a common environment, agents usually run decentralized TD($\lambda$) collaboratively. However, when some Byzantine agents behave adversarially, decentralized TD($\lambda$) is unable to learn an accurate linear approximation for the true value function. We propose a trimmed-mean based decentralized TD($\lambda$) algorithm to perform policy evaluation in this setting. We establish the finite-time convergence rate, as well as the asymptotic learning error that depends on the number of Byzantine agents. Numerical experiments corroborate the robustness of the proposed algorithm. |