Did I do that? Blame as a means to identify controlled effects in reinforcement learning

TL;DR

This paper introduces CEN, an unsupervised method that uses blame-based counterfactual measures to identify controllable effects in environments, improving exploration in reinforcement learning.

Contribution

The paper presents CEN, a novel unsupervised approach based on blame for identifying controlled effects, and demonstrates its effectiveness as an intrinsic motivator in reinforcement learning.

Findings

CEN accurately identifies controlled effects across various environments.

Integrating CEN with exploration methods significantly outperforms action-prediction models.

CEN enhances intrinsic motivation for reinforcement learning agents.

Abstract

Identifying controllable aspects of the environment has proven to be an extraordinary intrinsic motivator to reinforcement learning agents. Despite repeatedly achieving State-of-the-Art results, this approach has only been studied as a proxy to a reward-based task and has not yet been evaluated on its own. Current methods are based on action-prediction. Humans, on the other hand, assign blame to their actions to decide what they controlled. This work proposes Controlled Effect Network (CEN), an unsupervised method based on counterfactual measures of blame to identify effects on the environment controlled by the agent. CEN is evaluated in a wide range of environments showing that it can accurately identify controlled effects. Moreover, we demonstrate CEN's capabilities as intrinsic motivator by integrating it in the state-of-the-art exploration method, achieving substantially better performance than action-prediction models.