Fastest Convergence for Q-learning
This addresses faster and more stable convergence in reinforcement learning, particularly for Q-learning, but appears incremental as it builds on existing algorithms.
The paper introduces Zap Q-learning, an improved algorithm that achieves optimal asymptotic variance and matches deterministic Newton-Raphson behavior, with numerical experiments confirming quick convergence even in non-ideal settings.
The Zap Q-learning algorithm introduced in this paper is an improvement of Watkins' original algorithm and recent competitors in several respects. It is a matrix-gain algorithm designed so that its asymptotic variance is optimal. Moreover, an ODE analysis suggests that the transient behavior is a close match to a deterministic Newton-Raphson implementation. This is made possible by a two time-scale update equation for the matrix gain sequence. The analysis suggests that the approach will lead to stable and efficient computation even for non-ideal parameterized settings. Numerical experiments confirm the quick convergence, even in such non-ideal cases. A secondary goal of this paper is tutorial. The first half of the paper contains a survey on reinforcement learning algorithms, with a focus on minimum variance algorithms.