α-Nego: Self-Play Deep Reinforcement Learning for Negotiation Dialogues
Electronics · 11.05.2026 · 15 (10) · 2039
DOI: 10.3390/electronics15102039 ↗Kurz gefasst
Fachartikel in der Zeitschrift Electronics, online erschienen am 11.05.2026. Verfasst von Siqi Chen und 4 weiteren Autor:innen. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Negotiation is a complicated process that requires skills like strategic reasoning and communication. Most research aims at training dialogue agents for negotiation tasks using a few fixed opponents, which causes the agents to be effective only for these opponents and limits their strategy styles and performance across varying opponents. To yield better and more comprehensive strategies, we propose a novel self-play reinforcement learning (RL) framework for negotiation dialogues, named α-Nego, which allows one to train an RL agent against continuously improving opponents. For training, we introduce a holistic scoring approach that integrates utility with dialogue quality metrics (Agreement, Length, Social welfare), and we implement a tiered criterion for pool admission of selected opponents: utility dominance is primary, with holistic score components serving as deterministic tie-breakers to ensure selection pressure reflects both task success and dialogue quality. Furthermore, α-Nego uses a value distribution to enhance the ability of policy evaluation. This enables different styles of negotiation strategies to capture different risk attitudes by incorporating different criteria with a value distribution. Empirical evaluation on the Craigslistbargain and Dealornodeal dataset shows that the α-Nego agent clearly outperforms the state-of-the-art baselines.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Chen, S., Liu, J., Xiong, Z., Wang, Y., & Weiss, G. (2026). α-Nego: Self-Play Deep Reinforcement Learning for Negotiation Dialogues. Electronics, 15(10), 2039. https://doi.org/10.3390/electronics15102039
Vancouver
Chen S, Liu J, Xiong Z, Wang Y, Weiss G. α-Nego: Self-Play Deep Reinforcement Learning for Negotiation Dialogues. Electronics. 2026;15(10):2039. doi:10.3390/electronics15102039
BibTeX
@article{Chen2026,
author = {Chen, Siqi and Liu, Jinyi and Xiong, Zhaoyuan and Wang, Yunfei and Weiss, Gerhard},
title = {α-Nego: Self-Play Deep Reinforcement Learning for Negotiation Dialogues},
journal = {Electronics},
year = {2026},
volume = {15},
number = {10},
pages = {2039},
doi = {10.3390/electronics15102039}
}
EUVIMED