CoG 21

A research paper accepted at the IEEE Conference on Games 2021 that presents adversarial deep reinforcement learning for procedural content generation. It uses a Generator agent that creates challenges and a Solver agent that learns to solve them, with auxiliary inputs as a control variable for diversity.

Website Directory import
0 shouts Added Sep 23, 2026 ea.com
www.ea.com/seed/news/cog2021-adversarial-rl-content-generation
Preview unavailable

Screenshot unavailable. The listing stays available while a preview is prepared.

Links

Tags