Sample Efficient Reinforcement Learning via Large Vision Language Model Distillation

TL;DR

This paper introduces LVLM2P, a framework that distills large vision-language models into reinforcement learning agents, significantly improving sample efficiency and enabling more practical deployment in resource-constrained environments.

Contribution

The novel LVLM2P framework leverages large vision-language models as teachers to accelerate RL training and eliminate manual environment descriptions, enhancing efficiency and applicability.

Findings

LVLM2P significantly improves sample efficiency of RL algorithms.

The approach reduces early exploration inefficiencies.

It enables RL agents to operate without manual environment annotations.

Abstract

Recent research highlights the potential of multimodal foundation models in tackling complex decision-making challenges. However, their large parameters make real-world deployment resource-intensive and often impractical for constrained systems. Reinforcement learning (RL) shows promise for task-specific agents but suffers from high sample complexity, limiting practical applications. To address these challenges, we introduce LVLM to Policy (LVLM2P), a novel framework that distills knowledge from large vision-language models (LVLM) into more efficient RL agents. Our approach leverages the LVLM as a teacher, providing instructional actions based on trajectories collected by the RL agent, which helps reduce less meaningful exploration in the early stages of learning, thereby significantly accelerating the agent's learning progress. Additionally, by leveraging the LVLM to suggest actions directly from visual observations, we eliminate the need for manual textual descriptors of the environment, enhancing applicability across diverse tasks. Experiments show that LVLM2P significantly enhances the sample efficiency of baseline RL algorithms.