Implicitly Aligning Humans and Autonomous Agents through Shared Task Abstractions

2025/05/07 by Stéphane Aroca-Ouellette, Miguel Aroca-Ouellette, Aroca-Ouellette, Stéphane +5 · 1 citation
Computer Science · Psychology · #FOS: Computer and information sciences #Machine Learning (cs.LG) #Multiagent Systems (cs.MA) #Multimodal Machine Learning Applications #Reinforcement Learning in Robotics #Social Robot Interaction and HRI

paper · pdf · doi:10.48550/arxiv.2505.04579

openalex publication_date 2025/05/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

In collaborative tasks, autonomous agents fall short of humans in their capability to quickly adapt to new and unfamiliar teammates. We posit that a limiting factor for zero-shot coordination is the lack of shared task abstractions, a mechanism humans rely on to implicitly align with teammates. To address this gap, we introduce HA²: Hierarchical Ad Hoc Agents, a framework leveraging hierarchical reinforcement learning to mimic the structured approach humans use in collaboration. We evaluate HA² in the Overcooked environment, demonstrating statistically significant improvement over existing baselines when paired with both unseen agents and humans, providing better resilience to environmental shifts, and outperforming all state-of-the-art methods.

Citations

• Learning to Cooperate with Humans using Generative Agents
• Diverse Conventions for Human-AI Collaboration
• A Hierarchical Approach to Population Training for Human-AI Collaboration
• Voyager: An Open-Ended Embodied Agent with Large Language Models
• PECAN: Leveraging Policy Ensemble for Context-Aware Zero-Shot Human-AI Coordination
• Do As I Can, Not As I Say: Grounding Language in Robotic Affordances
• A Survey of Ad Hoc Teamwork Research
• Any-Play: An Intrinsic Augmentation for Zero-Shot Coordination
• Maximum Entropy Population-Based Training for Zero-Shot Human-AI Coordination
• Collaborating with Humans without Human Data
• Off-Belief Learning
• Open Problems in Cooperative AI
• RODE: Learning Roles to Decompose Multi-Agent Tasks
• Too many cooks: Bayesian inference for coordinating multi-agent collaboration
• "Other-Play" for Zero-Shot Coordination
• On the Utility of Learning about Humans for Human-AI Coordination
• Feudal Multi-Agent Hierarchies for Cooperative Reinforcement Learning
• FeUdal Networks for Hierarchical Reinforcement Learning
• CAPIR: Collaborative Action Planning with Intention Recognition
• Between MDPs and semi-MDPs: A framework for temporal abstraction in reinforcement learning
• Hierarchical Reinforcement Learning with the MAXQ Value Function Decomposition

Cited by

• Towards Zero-Shot Coordination between Teams of Agents: The N-XPlay Framework

Related

• Cross-environment Cooperation Enables Zero-shot Multi-agent Coordination
• A Hierarchical Approach to Population Training for Human-AI Collaboration
• TAG: A Decentralized Framework for Multi-Agent Hierarchical Reinforcement Learning
• Subgoal-based Hierarchical Reinforcement Learning for Multi-Agent Collaboration
• Designs for Enabling Collaboration in Human-Machine Teaming via Interactive and Explainable Systems