Effect of Private Deliberation: Deception of Large Language Models in Game Play

Author:

Poje Kristijan1,Brcic Mario1ORCID,Kovac Mihael1ORCID,Babac Marina Bagic1ORCID

Affiliation:

1. Faculty of Electrical Engineering and Computing, University of Zagreb, 10000 Zagreb, Croatia

Abstract

Integrating large language model (LLM) agents within game theory demonstrates their ability to replicate human-like behaviors through strategic decision making. In this paper, we introduce an augmented LLM agent, called the private agent, which engages in private deliberation and employs deception in repeated games. Utilizing the partially observable stochastic game (POSG) framework and incorporating in-context learning (ICL) and chain-of-thought (CoT) prompting, we investigated the private agent’s proficiency in both competitive and cooperative scenarios. Our empirical analysis demonstrated that the private agent consistently achieved higher long-term payoffs than its baseline counterpart and performed similarly or better in various game settings. However, we also found inherent deficiencies of LLMs in certain algorithmic capabilities crucial for high-quality decision making in games. These findings highlight the potential for enhancing LLM agents’ performance in multi-player games using information-theoretic approaches of deception and communication with complex environments.

Publisher

MDPI AG

Reference62 articles.

1. Language models are few-shot learners;Brown;Adv. Neural Inf. Process. Syst.,2020

2. Hoglund, S., and Khedri, J. (2024, May 01). Comparison Between RLHF and RLAIF in Fine-Tuning a Large Language Model. Available online: https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-331926.

3. Chain-of-thought prompting elicits reasoning in large language models;Wei;Adv. Neural Inf. Process. Syst.,2022

4. Creswell, A., Shanahan, M., and Higgins, I. (2022). Selection-inference: Exploiting large language models for interpretable logical reasoning. arXiv.

5. Meta Fundamental AI Research Diplomacy Team (FAIR), Bakhtin, A., Brown, N., Dinan, E., Farina, G., Flaherty, C., Fried, D., Goff, A., Gray, J., and Hu, H. (2022). Human-level play in the game of diplomacy by combining language models with strategic reasoning. Science, 378, 1067–1074.

Cited by 1 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3