ProAgent: Building Proactive Cooperative Agents with Large Language Models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zhang, Ceyao, Yang, Kaijie, Hu, Siyi, Wang, Zihao, Li, Guanghe, Sun, Yihang, Zhang, Cheng, Zhang, Zhaowei, Liu, Anji, Zhu, Song-Chun, Chang, Xiaojun, Zhang, Junge, Yin, Feng, Liang, Yitao, Yang, Yaodong
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914637748895744
author Zhang, Ceyao
Yang, Kaijie
Hu, Siyi
Wang, Zihao
Li, Guanghe
Sun, Yihang
Zhang, Cheng
Zhang, Zhaowei
Liu, Anji
Zhu, Song-Chun
Chang, Xiaojun
Zhang, Junge
Yin, Feng
Liang, Yitao
Yang, Yaodong
author_facet Zhang, Ceyao
Yang, Kaijie
Hu, Siyi
Wang, Zihao
Li, Guanghe
Sun, Yihang
Zhang, Cheng
Zhang, Zhaowei
Liu, Anji
Zhu, Song-Chun
Chang, Xiaojun
Zhang, Junge
Yin, Feng
Liang, Yitao
Yang, Yaodong
contents Building agents with adaptive behavior in cooperative tasks stands as a paramount goal in the realm of multi-agent systems. Current approaches to developing cooperative agents rely primarily on learning-based methods, whose policy generalization depends heavily on the diversity of teammates they interact with during the training phase. Such reliance, however, constrains the agents' capacity for strategic adaptation when cooperating with unfamiliar teammates, which becomes a significant challenge in zero-shot coordination scenarios. To address this challenge, we propose ProAgent, a novel framework that harnesses large language models (LLMs) to create proactive agents capable of dynamically adapting their behavior to enhance cooperation with teammates. ProAgent can analyze the present state, and infer the intentions of teammates from observations. It then updates its beliefs in alignment with the teammates' subsequent actual behaviors. Moreover, ProAgent exhibits a high degree of modularity and interpretability, making it easily integrated into various of coordination scenarios. Experimental evaluations conducted within the Overcooked-AI environment unveil the remarkable performance superiority of ProAgent, outperforming five methods based on self-play and population-based training when cooperating with AI agents. Furthermore, in partnered with human proxy models, its performance exhibits an average improvement exceeding 10% compared to the current state-of-the-art method. For more information about our project, please visit~\url{https://pku-proagent.github.io}.
format Preprint
id arxiv_https___arxiv_org_abs_2308_11339
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle ProAgent: Building Proactive Cooperative Agents with Large Language Models
Zhang, Ceyao
Yang, Kaijie
Hu, Siyi
Wang, Zihao
Li, Guanghe
Sun, Yihang
Zhang, Cheng
Zhang, Zhaowei
Liu, Anji
Zhu, Song-Chun
Chang, Xiaojun
Zhang, Junge
Yin, Feng
Liang, Yitao
Yang, Yaodong
Artificial Intelligence
Machine Learning
Multiagent Systems
Building agents with adaptive behavior in cooperative tasks stands as a paramount goal in the realm of multi-agent systems. Current approaches to developing cooperative agents rely primarily on learning-based methods, whose policy generalization depends heavily on the diversity of teammates they interact with during the training phase. Such reliance, however, constrains the agents' capacity for strategic adaptation when cooperating with unfamiliar teammates, which becomes a significant challenge in zero-shot coordination scenarios. To address this challenge, we propose ProAgent, a novel framework that harnesses large language models (LLMs) to create proactive agents capable of dynamically adapting their behavior to enhance cooperation with teammates. ProAgent can analyze the present state, and infer the intentions of teammates from observations. It then updates its beliefs in alignment with the teammates' subsequent actual behaviors. Moreover, ProAgent exhibits a high degree of modularity and interpretability, making it easily integrated into various of coordination scenarios. Experimental evaluations conducted within the Overcooked-AI environment unveil the remarkable performance superiority of ProAgent, outperforming five methods based on self-play and population-based training when cooperating with AI agents. Furthermore, in partnered with human proxy models, its performance exhibits an average improvement exceeding 10% compared to the current state-of-the-art method. For more information about our project, please visit~\url{https://pku-proagent.github.io}.
title ProAgent: Building Proactive Cooperative Agents with Large Language Models
topic Artificial Intelligence
Machine Learning
Multiagent Systems
url https://arxiv.org/abs/2308.11339