Self-imitation Learning for Action Generation in Text-based Games

Publication Type:
Conference Proceeding
Citation:
EACL 2023 - 17th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference, 2023, pp. 703-726
Issue Date:
2023-01-01
Filename Description Size
2023.eacl-main.50.pdfPublished version4.1 MB
Adobe PDF
Full metadata record
In this work, we study reinforcement learning (RL) in solving text-based games. We address the challenge of combinatorial action space, by proposing a confidence-based self-imitation model to generate action candidates for the RL agent. Firstly, we leverage the self-imitation learning to rank and exploit past valuable trajectories to adapt a pre-trained language model (LM) towards a target game. Then, we devise a confidence-based strategy to measure the LM's confidence with respect to a state, thus adaptively pruning the generated actions to yield a more compact set of action candidates. In multiple challenging games, our model demonstrates promising performance in comparison to the baselines.
Please use this identifier to cite or link to this item: