An Exact Distributed Newton Method for Reinforcement Learning

TL;DR

This paper introduces a novel distributed second-order method for reinforcement learning that outperforms existing algorithms in speed by leveraging sparsity and solving specialized linear systems, with proven super-linear convergence.

Contribution

The paper presents the first exact distributed Newton method for reinforcement learning that exploits dual Hessian sparsity for faster convergence and improved performance.

Findings

Outperforms state-of-the-art methods including ADMM

Proven super-linear convergence near the optimal solution

Demonstrated superior empirical performance on benchmark tasks

Abstract

In this paper, we propose a distributed second- order method for reinforcement learning. Our approach is the fastest in literature so-far as it outperforms state-of-the-art methods, including ADMM, by significant margins. We achieve this by exploiting the sparsity pattern of the dual Hessian and transforming the problem of computing the Newton direction to one of solving a sequence of symmetric diagonally dominant system of equations. We validate the above claim both theoretically and empirically. On the theoretical side, we prove that similar to exact Newton, our algorithm exhibits super-linear convergence within a neighborhood of the optimal solution. Empirically, we demonstrate the superiority of this new method on a set of benchmark reinforcement learning tasks.