Deep Reinforcement Learning for Game Testing at EA with Konrad Tollmar - #517
Published:Sep 9, 2021 17:35
•1 min read
•Practical AI
Analysis
This article from Practical AI discusses the application of deep reinforcement learning (DRL) in game testing at Electronic Arts (EA). It features an interview with Konrad Tollmar, a research director at EA and professor at KTH, focusing on how EA's SEED team utilizes ML/AI in popular game franchises like Apex Legends, Madden, and FIFA. The conversation covers the team's research agenda, the challenges of applying DRL to modern 3D games compared to Atari games, the use of Convolutional Neural Networks (CNNs) for glitch detection, and Tollmar's perspective on the future of ML in game training. The article highlights the practical applications of AI in the gaming industry.
Key Takeaways
Reference
“We break down a few papers focused on the application of ML to game testing, discussing why deep reinforcement learning is at the top of their research agenda...”