CAMEO: Curiosity Augmented Metropolis for Exploratory Optimal Policies

TL;DR

This paper introduces CAMEO, a curiosity-augmented Metropolis algorithm that samples diverse optimal policies in reinforcement learning, enabling better coverage of behaviors and risk profiles in control tasks.

Contribution

The paper proposes a novel method, CAMEO, to sample and analyze the distribution of optimal policies, enhancing understanding of policy diversity in reinforcement learning.

Findings

CAMEO successfully samples diverse optimal policies in classic control problems.

The method handles environments with sparse rewards effectively.

Sampled policies exhibit different risk profiles, useful for interpretability.

Abstract

Reinforcement Learning has drawn huge interest as a tool for solving optimal control problems. Solving a given problem (task or environment) involves converging towards an optimal policy. However, there might exist multiple optimal policies that can dramatically differ in their behaviour; for example, some may be faster than the others but at the expense of greater risk. We consider and study a distribution of optimal policies. We design a curiosity-augmented Metropolis algorithm (CAMEO), such that we can sample optimal policies, and such that these policies effectively adopt diverse behaviours, since this implies greater coverage of the different possible optimal policies. In experimental simulations we show that CAMEO indeed obtains policies that all solve classic control problems, and even in the challenging case of environments that provide sparse rewards. We further show that the different policies we sample present different risk profiles, corresponding to interesting practical applications in interpretability, and represents a first step towards learning the distribution of optimal policies itself.