On the Critical Role of Conventions in Adaptive Human-AI Collaboration

TL;DR

This paper introduces a learning framework that separates rule-based and convention-based representations to enable AI agents to adapt quickly to new human partners and tasks in collaborative settings, mimicking human adaptability.

Contribution

The work presents a novel approach to distinguish and leverage conventions in AI adaptation, improving zero-shot coordination in diverse collaborative tasks.

Findings

Agents adapt quickly to new partners and tasks.

Separation of rule and convention representations enhances flexibility.

Validated on three diverse collaborative tasks.

Abstract

Humans can quickly adapt to new partners in collaborative tasks (e.g. playing basketball), because they understand which fundamental skills of the task (e.g. how to dribble, how to shoot) carry over across new partners. Humans can also quickly adapt to similar tasks with the same partners by carrying over conventions that they have developed (e.g. raising hand signals pass the ball), without learning to coordinate from scratch. To collaborate seamlessly with humans, AI agents should adapt quickly to new partners and new tasks as well. However, current approaches have not attempted to distinguish between the complexities intrinsic to a task and the conventions used by a partner, and more generally there has been little focus on leveraging conventions for adapting to new settings. In this work, we propose a learning framework that teases apart rule-dependent representation from convention-dependent representation in a principled way. We show that, under some assumptions, our rule-dependent representation is a sufficient statistic of the distribution over best-response strategies across partners. Using this separation of representations, our agents are able to adapt quickly to new partners, and to coordinate with old partners on new tasks in a zero-shot manner. We experimentally validate our approach on three collaborative tasks varying in complexity: a contextual multi-armed bandit, a block placing task, and the card game Hanabi.