LaGR-SEQ: Language-Guided Reinforcement Learning with Sample-Efficient Querying
This work addresses sample inefficiency in RL training for researchers, though it is incremental as it builds on existing LLM and RL methods.
The paper tackles the problem of inefficient reinforcement learning (RL) training by using large language models (LLMs) to propose solutions for partially completed tasks, but addresses the high cost of frequent LLM queries with a secondary RL agent that decides when to query. The result is a framework, LaGR-SEQ, that enables more efficient primary RL training while minimizing LLM queries, demonstrated on a series of tasks.
Large language models (LLMs) have recently demonstrated their impressive ability to provide context-aware responses via text. This ability could potentially be used to predict plausible solutions in sequential decision making tasks pertaining to pattern completion. For example, by observing a partial stack of cubes, LLMs can predict the correct sequence in which the remaining cubes should be stacked by extrapolating the observed patterns (e.g., cube sizes, colors or other attributes) in the partial stack. In this work, we introduce LaGR (Language-Guided Reinforcement learning), which uses this predictive ability of LLMs to propose solutions to tasks that have been partially completed by a primary reinforcement learning (RL) agent, in order to subsequently guide the latter's training. However, as RL training is generally not sample-efficient, deploying this approach would inherently imply that the LLM be repeatedly queried for solutions; a process that can be expensive and infeasible. To address this issue, we introduce SEQ (sample efficient querying), where we simultaneously train a secondary RL agent to decide when the LLM should be queried for solutions. Specifically, we use the quality of the solutions emanating from the LLM as the reward to train this agent. We show that our proposed framework LaGR-SEQ enables more efficient primary RL training, while simultaneously minimizing the number of queries to the LLM. We demonstrate our approach on a series of tasks and highlight the advantages of our approach, along with its limitations and potential future research directions.