Dialogue Shaping: Empowering Agents through NPC Interaction
This addresses the problem of inefficient training in text-based game environments for reinforcement learning practitioners, representing an incremental improvement.
The paper tackles the challenge of slow convergence in reinforcement learning for text-based games by using large language models to extract key information from non-player characters and incorporating it via knowledge graphs and Story Shaping, resulting in faster training.
One major challenge in reinforcement learning (RL) is the large amount of steps for the RL agent needs to converge in the training process and learn the optimal policy, especially in text-based game environments where the action space is extensive. However, non-player characters (NPCs) sometimes hold some key information about the game, which can potentially help to train RL agents faster. Thus, this paper explores how to interact and converse with NPC agents to get the key information using large language models (LLMs), as well as incorporate this information to speed up RL agent's training using knowledge graphs (KGs) and Story Shaping.