UCB Momentum Q-learning: Correcting the bias without forgetting

by   Pierre Ménard, et al.

We propose UCBMQ, Upper Confidence Bound Momentum Q-learning, a new algorithm for reinforcement learning in tabular and possibly stage-dependent, episodic Markov decision process. UCBMQ is based on Q-learning where we add a momentum term and rely on the principle of optimism in face of uncertainty to deal with exploration. Our new technical ingredient of UCBMQ is the use of momentum to correct the bias that Q-learning suffers while, at the same time, limiting the impact it has on the second-order term of the regret. For UCBMQ , we are able to guarantee a regret of at most O(√(H^3SAT)+ H^4 S A ) where H is the length of an episode, S the number of states, A the number of actions, T the number of episodes and ignoring terms in polylog(SAHT). Notably, UCBMQ is the first algorithm that simultaneously matches the lower bound of Ω(√(H^3SAT)) for large enough T and has a second-order term (with respect to the horizon T) that scales only linearly with the number of states S.


page 1

page 2

page 3

page 4


Optimistic Posterior Sampling for Reinforcement Learning with Few Samples and Tight Guarantees

We consider reinforcement learning in an environment modeled by an episo...

Provably Efficient Reinforcement Learning with Aggregated States

We establish that an optimistic variant of Q-learning applied to a finit...

From Dirichlet to Rubin: Optimistic Exploration in RL without Bonuses

We propose the Bayes-UCBVI algorithm for reinforcement learning in tabul...

Fine-Grained Gap-Dependent Bounds for Tabular MDPs via Adaptive Multi-Step Bootstrap

This paper presents a new model-free algorithm for episodic finite-horiz...

Double Thompson Sampling in Finite stochastic Games

We consider the trade-off problem between exploration and exploitation u...

Online Sparse Reinforcement Learning

We investigate the hardness of online reinforcement learning in fixed ho...

Correcting Momentum in Temporal Difference Learning

A common optimization tool used in deep reinforcement learning is moment...

Please sign up or login with your details

Forgot password? Click here to reset