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.
Adithya M Devraj (University of Florida)
Sean Meyn (University of Florida)
More from the Same Authors
2019 Poster: Stochastic Variance Reduced Primal Dual Algorithms for Empirical Composition Optimization »
Adithya M Devraj · Jianshu Chen