Posterior Sampling for Continuing Environments
We develop an extension of posterior sampling for reinforcement learning (PSRL) that is suited for a continuing agent-environment interface and integrates naturally into agent designs that scale to complex environments. The approach maintains a statistically plausible model of the environment and follows a policy that maximizes expected γ-discounted return in that model. At each time, with probability 1-γ, the model is replaced by a sample from the posterior distribution over environments. For a suitable schedule of γ, we establish an Õ(τ S √(A T)) bound on the Bayesian regret, where S is the number of environment states, A is the number of actions, and τ denotes the reward averaging time, which is a bound on the duration required to accurately estimate the average reward of any policy.
READ FULL TEXT