Incorporating Rivalry in Reinforcement Learning for a Competitive Game

TL;DR

This paper introduces a novel reinforcement learning approach incorporating rivalry to enhance social impact in competitive games, aiming to improve human perception of AI agents.

Contribution

It proposes a new rivalry-based learning mechanism for reinforcement learning agents in competitive scenarios involving humans.

Findings

Rivalry influences human perception of AI agents.

Agents with rivalry-based learning perform competitively.

The social impact of agents can be modulated through rivalry.

Abstract

Recent advances in reinforcement learning with social agents have allowed us to achieve human-level performance on some interaction tasks. However, most interactive scenarios do not have as end-goal performance alone; instead, the social impact of these agents when interacting with humans is as important and, in most cases, never explored properly. This preregistration study focuses on providing a novel learning mechanism based on a rivalry social impact. Our scenario explored different reinforcement learning-based agents playing a competitive card game against human players. Based on the concept of competitive rivalry, our analysis aims to investigate if we can change the assessment of these agents from a human perspective.