A Study of Policy Gradient on a Class of Exactly Solvable Models

Policy gradient methods are extensively used in reinforcement learning as a way to optimize expected return. In this paper, we explore the evolution of the policy parameters, for a special class of exactly solvable POMDPs, as a continuous-state Markov chain, whose transition probabilities are determ...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2020-11
Main Authors: McCracken, Gavin, Daniels, Colin, Zhao, Rosie, Brandenberger, Anna, Panangaden, Prakash, Precup, Doina
Format: Article
Language:English
Subjects:
Evolution
Markov chains
Maxima
Parameters
Random walk
Random walk theory
Transition probabilities
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Policy gradient methods are extensively used in reinforcement learning as a way to optimize expected return. In this paper, we explore the evolution of the policy parameters, for a special class of exactly solvable POMDPs, as a continuous-state Markov chain, whose transition probabilities are determined by the gradient of the distribution of the policy's value. Our approach relies heavily on random walk theory, specifically on affine Weyl groups. We construct a class of novel partially observable environments with controllable exploration difficulty, in which the value distribution, and hence the policy parameter evolution, can be derived analytically. Using these environments, we analyze the probabilistic convergence of policy gradient to different local maxima of the value function. To our knowledge, this is the first approach developed to analytically compute the landscape of policy gradient in POMDPs for a class of such environments, leading to interesting insights into the difficulty of this problem.
ISSN:2331-8422