Fast Rates for the Regret of Offline Reinforcement Learning: Under the Margin Condition, the Greedy Policy of a Q*-Estimate with Uniform Sub-Gaussian Error at Rate aₙ Has Regret O(aₙ^(1+α))Research Paper
Offline reinforcement learning asks for a good policy in a Markov decision process using only logged transitions, without further interaction. A common family of methods (fitted -iteration, Bellman residual minimization) first estimates the optimal action-value function by some and then acts greedily with respect to . Standard analyses…