Ad-Hoc Human-AI Coordination Challenge

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Dizdarević, Tin, Hammond, Ravi, Gessler, Tobias, Calinescu, Anisoara, Cook, Jonathan, Gallici, Matteo, Lupu, Andrei, Muglich, Darius, Forkel, Johannes, Foerster, Jakob Nicolaus
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912455668531200
author Dizdarević, Tin
Hammond, Ravi
Gessler, Tobias
Calinescu, Anisoara
Cook, Jonathan
Gallici, Matteo
Lupu, Andrei
Muglich, Darius
Forkel, Johannes
Foerster, Jakob Nicolaus
author_facet Dizdarević, Tin
Hammond, Ravi
Gessler, Tobias
Calinescu, Anisoara
Cook, Jonathan
Gallici, Matteo
Lupu, Andrei
Muglich, Darius
Forkel, Johannes
Foerster, Jakob Nicolaus
contents Achieving seamless coordination between AI agents and humans is crucial for real-world applications, yet it remains a significant open challenge. Hanabi is a cooperative card game featuring imperfect information, constrained communication, theory of mind requirements, and coordinated action -- making it an ideal testbed for human-AI coordination. However, its use for human-AI interaction has been limited by the challenges of human evaluation. In this work, we introduce the Ad-Hoc Human-AI Coordination Challenge (AH2AC2) to overcome the constraints of costly and difficult-to-reproduce human evaluations. We develop \textit{human proxy agents} on a large-scale human dataset that serve as robust, cheap, and reproducible human-like evaluation partners in AH2AC2. To encourage the development of data-efficient methods, we open-source a dataset of 3,079 games, deliberately limiting the amount of available human gameplay data. We present baseline results for both two- and three- player Hanabi scenarios. To ensure fair evaluation, we host the proxy agents through a controlled evaluation system rather than releasing them publicly. The code is available at \href{https://github.com/FLAIROx/ah2ac2}{https://github.com/FLAIROx/ah2ac2}.
format Preprint
id arxiv_https___arxiv_org_abs_2506_21490
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Ad-Hoc Human-AI Coordination Challenge
Dizdarević, Tin
Hammond, Ravi
Gessler, Tobias
Calinescu, Anisoara
Cook, Jonathan
Gallici, Matteo
Lupu, Andrei
Muglich, Darius
Forkel, Johannes
Foerster, Jakob Nicolaus
Artificial Intelligence
Human-Computer Interaction
Multiagent Systems
Achieving seamless coordination between AI agents and humans is crucial for real-world applications, yet it remains a significant open challenge. Hanabi is a cooperative card game featuring imperfect information, constrained communication, theory of mind requirements, and coordinated action -- making it an ideal testbed for human-AI coordination. However, its use for human-AI interaction has been limited by the challenges of human evaluation. In this work, we introduce the Ad-Hoc Human-AI Coordination Challenge (AH2AC2) to overcome the constraints of costly and difficult-to-reproduce human evaluations. We develop \textit{human proxy agents} on a large-scale human dataset that serve as robust, cheap, and reproducible human-like evaluation partners in AH2AC2. To encourage the development of data-efficient methods, we open-source a dataset of 3,079 games, deliberately limiting the amount of available human gameplay data. We present baseline results for both two- and three- player Hanabi scenarios. To ensure fair evaluation, we host the proxy agents through a controlled evaluation system rather than releasing them publicly. The code is available at \href{https://github.com/FLAIROx/ah2ac2}{https://github.com/FLAIROx/ah2ac2}.
title Ad-Hoc Human-AI Coordination Challenge
topic Artificial Intelligence
Human-Computer Interaction
Multiagent Systems
url https://arxiv.org/abs/2506.21490