Selective Dyna-style Planning Under Limited Model Capacity

TL;DR

This paper explores selective planning in model-based reinforcement learning by combining uncertainty estimates from heteroscedastic regression and parameter uncertainty to improve planning effectiveness with imperfect models.

Contribution

It introduces a method to estimate predictive uncertainty from model inadequacy using heteroscedastic regression, complementing existing parameter uncertainty approaches for selective planning.

Findings

Heteroscedastic regression signals model inadequacy effectively.

Combining uncertainty sources improves selective planning.

Model inadequacy detection complements parameter uncertainty methods.

Abstract

In model-based reinforcement learning, planning with an imperfect model of the environment has the potential to harm learning progress. But even when a model is imperfect, it may still contain information that is useful for planning. In this paper, we investigate the idea of using an imperfect model selectively. The agent should plan in parts of the state space where the model would be helpful but refrain from using the model where it would be harmful. An effective selective planning mechanism requires estimating predictive uncertainty, which arises out of aleatoric uncertainty, parameter uncertainty, and model inadequacy, among other sources. Prior work has focused on parameter uncertainty for selective planning. In this work, we emphasize the importance of model inadequacy. We show that heteroscedastic regression can signal predictive uncertainty arising from model inadequacy that is complementary to that which is detected by methods designed for parameter uncertainty, indicating that considering both parameter uncertainty and model inadequacy may be a more promising direction for effective selective planning than either in isolation.