Can Q-Learning be Improved with Advice?
This work extends algorithms with predictions from simple online problems to reinforcement learning, offering potential improvements for RL practitioners dealing with structured MDPs.
The paper tackles the problem of improving Q-learning regret bounds by incorporating prior predictions about the optimal Q-value function, showing that under a weak condition called distillation, regret can be reduced by focusing on inaccurate predictions, with sublinear regret even for arbitrary predictions.
Despite rapid progress in theoretical reinforcement learning (RL) over the last few years, most of the known guarantees are worst-case in nature, failing to take advantage of structure that may be known a priori about a given RL problem at hand. In this paper we address the question of whether worst-case lower bounds for regret in online learning of Markov decision processes (MDPs) can be circumvented when information about the MDP, in the form of predictions about its optimal $Q$-value function, is given to the algorithm. We show that when the predictions about the optimal $Q$-value function satisfy a reasonably weak condition we call distillation, then we can improve regret bounds by replacing the set of state-action pairs with the set of state-action pairs on which the predictions are grossly inaccurate. This improvement holds for both uniform regret bounds and gap-based ones. Further, we are able to achieve this property with an algorithm that achieves sublinear regret when given arbitrary predictions (i.e., even those which are not a distillation). Our work extends a recent line of work on algorithms with predictions, which has typically focused on simple online problems such as caching and scheduling, to the more complex and general problem of reinforcement learning.