Searching for Plannable Domains can Speed up Reinforcement Learning

TL;DR

This paper introduces plannable RL (pRL), a method that identifies near-deterministic domains to focus planning, enabling faster learning and near-optimal macro actions in reinforcement learning tasks.

Contribution

The paper proposes pRL, a novel approach that separates plannable domains to improve planning efficiency and speed up reinforcement learning.

Findings

pRL finds an optimal policy in tested environments.

Plannable macro actions are near-optimal.

pRL enables faster learning compared to traditional methods.

Abstract

Reinforcement learning (RL) involves sequential decision making in uncertain environments. The aim of the decision-making agent is to maximize the benefit of acting in its environment over an extended period of time. Finding an optimal policy in RL may be very slow. To speed up learning, one often used solution is the integration of planning, for example, Sutton's Dyna algorithm, or various other methods using macro-actions. Here we suggest to separate plannable, i.e., close to deterministic parts of the world, and focus planning efforts in this domain. A novel reinforcement learning method called plannable RL (pRL) is proposed here. pRL builds a simple model, which is used to search for macro actions. The simplicity of the model makes planning computationally inexpensive. It is shown that pRL finds an optimal policy, and that plannable macro actions found by pRL are near-optimal. In turn, it is unnecessary to try large numbers of macro actions, which enables fast learning. The utility of pRL is demonstrated by computer simulations.