Result: pgeon applied to Overcooked-AI to explain agents' behaviour

Title:
pgeon applied to Overcooked-AI to explain agents' behaviour
Authors:
Universitat Politècnica de Catalunya. Doctorat en Intel·ligència Artificial, Universitat Politècnica de Catalunya. Departament de Ciències de la Computació, Universitat Politècnica de Catalunya. IDEAI-UPC - Intelligent Data sciEnce and Artificial Intelligence Research Group, Tormos Llorente, Adrián, Giménez Ábalos, Víctor, Vázquez Salceda, Javier, Álvarez Napagao, Sergio
Publisher Information:
Association for Computing Machinery (ACM) 2024
Document Type:
Electronic Resource Electronic Resource
Index Terms:
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Agents intel·ligents, Explainability, Multi-agent systems, Reinforcement learning, Autonomous agents, Action spaces, Agent behavior, AI systems, Cooperative environment, Human-systems, State-space, Multi agent systems, Conference report
URL:
http://hdl.handle.net/2117/419900
https://dl.acm.org/doi/abs/10.5555/3635637.3663299
Availability:
Open access content. Open access content
Attribution 4.0 International
http://creativecommons.org/licenses/by/4.0
Open Access
Note:
3 p.
application/pdf
English
Other Numbers:
HGF oai:upcommons.upc.edu:2117/419900
Tormos, A. [et al.]. pgeon applied to Overcooked-AI to explain agents' behaviour. A: International Conference on Autonomous Agents and Multiagent Systems. "AAMAS '24: Proceedings of the 23rd International Conference on Autonomous Agents and Multiagent Systems". New York: Association for Computing Machinery (ACM), 2024, p. 2821-2823. ISBN 979-8-4007-0486-4. DOI 10.5555/3635637.3663299 .
979-8-4007-0486-4
10.5555/3635637.3663299
1479311884
Contributing Source:
UNIV POLITECNICA DE CATALUNYA
From OAIster®, provided by the OCLC Cooperative.
Accession Number:
edsoai.on1479311884
Database:
OAIster

Further Information

Policy Graphs (PGs) are a method for representing the behaviour of opaque agents by observing them in the environment and producing graphs where the state and action spaces are discretised into predicates. We present pgeon, a Python library that demonstrates the effectiveness of PGs in providing explanations for the behaviour of agents and we showcase it by applying it to a multi-agent cooperative environment: Overcooked-AI. This library illustrates how PGs can create transparent and explainable surrogate agents that closely mimic the behavior of the original agents. These features can help improving trust in environments where humans and AI systems collaborate by improving the explainability of all agents, even opaque or human. © 2024 International Foundation for Autonomous Agents and Multiagent Systems.
This work has been partially supported by the AI4CCAM (Grant agreement ID: 101076911) and the HumanE-AI-Net (Grant agreement ID: 952026) European projects.
Peer Reviewed
Postprint (published version)