Tell me why! Explanations support learning relational and causal structure
This addresses the problem of improving agent learning and generalization in AI, particularly for reinforcement-learning agents, by leveraging language as a tool, representing a novel application rather than an incremental step.
The paper tackles the challenge of reinforcement-learning agents inferring relational and causal structure by showing that training agents to predict language descriptions and explanations improves their performance in complex environments, enabling them to learn relational tasks, generalize out-of-distribution, and perform experimental interventions to identify causal relationships.
Inferring the abstract relational and causal structure of the world is a major challenge for reinforcement-learning (RL) agents. For humans, language--particularly in the form of explanations--plays a considerable role in overcoming this challenge. Here, we show that language can play a similar role for deep RL agents in complex environments. While agents typically struggle to acquire relational and causal knowledge, augmenting their experience by training them to predict language descriptions and explanations can overcome these limitations. We show that language can help agents learn challenging relational tasks, and examine which aspects of language contribute to its benefits. We then show that explanations can help agents to infer not only relational but also causal structure. Language can shape the way that agents to generalize out-of-distribution from ambiguous, causally-confounded training, and explanations even allow agents to learn to perform experimental interventions to identify causal relationships. Our results suggest that language description and explanation may be powerful tools for improving agent learning and generalization.