OCALM: Object-Centric Assessment with Language Models

TL;DR

OCALM introduces an interpretable reward framework for reinforcement learning that leverages language models and object-centric environment understanding to facilitate goal specification and policy derivation from natural language descriptions.

Contribution

This work presents OCALM, a novel method combining language models and object-centric reasoning to generate transparent reward functions from natural language tasks.

Findings

OCALM produces interpretable reward functions.

It enables RL agents to learn from natural language descriptions.

The approach leverages object-centric environment understanding.

Abstract

Properly defining a reward signal to efficiently train a reinforcement learning (RL) agent is a challenging task. Designing balanced objective functions from which a desired behavior can emerge requires expert knowledge, especially for complex environments. Learning rewards from human feedback or using large language models (LLMs) to directly provide rewards are promising alternatives, allowing non-experts to specify goals for the agent. However, black-box reward models make it difficult to debug the reward. In this work, we propose Object-Centric Assessment with Language Models (OCALM) to derive inherently interpretable reward functions for RL agents from natural language task descriptions. OCALM uses the extensive world-knowledge of LLMs while leveraging the object-centric nature common to many environments to derive reward functions focused on relational concepts, providing RL agents with the ability to derive policies from task descriptions.