Technical Challenges of Deploying Reinforcement Learning Agents for Game Testing in AAA Games

TL;DR

This paper discusses integrating reinforcement learning into automated game testing for AAA titles, highlighting challenges, implementation details, and proposing future research directions to facilitate industry adoption.

Contribution

It presents a case study of applying reinforcement learning to enhance game testing coverage in AAA games, addressing deployment challenges and suggesting research pathways.

Findings

Reinforcement learning increased test coverage in AAA games.

Integration posed significant technical challenges.

Proposed research directions to improve RL deployment in game testing.

Abstract

Going from research to production, especially for large and complex software systems, is fundamentally a hard problem. In large-scale game production, one of the main reasons is that the development environment can be very different from the final product. In this technical paper we describe an effort to add an experimental reinforcement learning system to an existing automated game testing solution based on scripted bots in order to increase its capacity. We report on how this reinforcement learning system was integrated with the aim to increase test coverage similar to [1] in a set of AAA games including Battlefield 2042 and Dead Space (2023). The aim of this technical paper is to show a use-case of leveraging reinforcement learning in game production and cover some of the largest time sinks anyone who wants to make the same journey for their game may encounter. Furthermore, to help the game industry to adopt this technology faster, we propose a few research directions that we believe will be valuable and necessary for making machine learning, and especially reinforcement learning, an effective tool in game production.