Convergence and Regret of the Policy Gradient for Multi-Armed Bandits in Diffusion Environment

Part of

Advanced Temporal And Weighted Algorithms