Abstract
Despite recent advancements in AI and NLP, negotiation remains a difficult domain for AI agents. Traditional game theoretic approaches that have worked well for two-player zero-sum games struggle in the context of negotiation due to their inability to learn human-compatible strategies. On the other hand, approaches that only use human data tend to be domain-specific and lack the theoretical guarantees provided by strategies grounded in game theory. Motivated by the notion of fairness as a criterion for optimality in general sum games, we propose a negotiation framework called FDHC which incorporates fairness into both the reward design and search to learn human-compatible negotiation strategies. Our method includes a novel, RL+search technique called LGM-Zero which leverages a pre-trained language model to retrieve human-compatible offers from large action spaces. Our results show that our method is able to achieve more egalitarian negotiation outcomes and improve negotiation quality.- Anthology ID:
- 2024.findings-emnlp.308
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2024
- Month:
- November
- Year:
- 2024
- Address:
- Miami, Florida, USA
- Editors:
- Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 5346–5370
- Language:
- URL:
- https://aclanthology.org/2024.findings-emnlp.308
- DOI:
- 10.18653/v1/2024.findings-emnlp.308
- Cite (ACL):
- Ryan Shea and Zhou Yu. 2024. A Fairness-Driven Method for Learning Human-Compatible Negotiation Strategies. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 5346–5370, Miami, Florida, USA. Association for Computational Linguistics.
- Cite (Informal):
- A Fairness-Driven Method for Learning Human-Compatible Negotiation Strategies (Shea & Yu, Findings 2024)
- PDF:
- https://preview.aclanthology.org/dois-2013-emnlp/2024.findings-emnlp.308.pdf