Hidden Agenda: a Social Deduction Game with Diverse Learned Equilibria

TL;DR

This paper introduces Hidden Agenda, a social deduction game environment for studying how learning agents can develop diverse strategies and cooperation behaviors in scenarios with hidden motivations and unknown team alignments.

Contribution

It presents a novel 2D environment for multiagent social deduction, enabling analysis of diverse learned equilibria and strategic behaviors without communication.

Findings

Agents learn various cooperation and voting strategies

Agents operate effectively without natural language communication

Environment supports rich strategic diversity

Abstract

A key challenge in the study of multiagent cooperation is the need for individual agents not only to cooperate effectively, but to decide with whom to cooperate. This is particularly critical in situations when other agents have hidden, possibly misaligned motivations and goals. Social deduction games offer an avenue to study how individuals might learn to synthesize potentially unreliable information about others, and elucidate their true motivations. In this work, we present Hidden Agenda, a two-team social deduction game that provides a 2D environment for studying learning agents in scenarios of unknown team alignment. The environment admits a rich set of strategies for both teams. Reinforcement learning agents trained in Hidden Agenda show that agents can learn a variety of behaviors, including partnering and voting without need for communication in natural language.