Auto MC-Reward: Automated Dense Reward Design with Large Language Models for Minecraft

arXiv:2312.09238v2 Announce Type: replace-cross
Abstract: Many reinforcement learning environments (e.g., Minecraft) provide only sparse rewards that indicate task completion or failure with binary values. The challenge in exploration efficiency in such environments makes it difficult for reinforcement-learning-based agents to learn complex tasks. To address this, this paper introduces an advanced learning system, named Auto MC-Reward, that leverages Large Language Models (LLMs) to automatically design dense reward functions, thereby enhancing the learning efficiency. Auto MC-Reward consists of three important components: Reward …

abstract agents arxiv auto automated binary challenge cs.ai cs.cl cs.cv cs.lg design efficiency environments exploration failure language language models large language large language models learn minecraft paper reinforcement reinforcement learning tasks type values