Computer Science38th Conference on Neural Information Processing Systems (NeurIPS 2024)
Diffusion for World Modeling: Visual Details Matter in Atari
E. Alonso, A. Jelley, et al.
DIAMOND trains reinforcement learning agents inside a diffusion-based world model to capture richer visual details and improve agent performance, achieving a mean human normalized score of 1.46 on the Atari 100k benchmark and demonstrating a playable neural game engine on Counter-Strike: Global Offensive — research conducted by Eloi Alonso, Adam Jelley, Vincent Micheli, Anssi Kanervisto, Amos Storkey, Tim Pearce, François Fleuret.
Related Publications
Explore these studies to deepen your understanding
Adjacent work that informs or extends this paper's methodology and findings.
Computer Science
Diffusion for World Modeling: Visual Details Matter in Atari
E. Alonso, A. Jelley, et al.
Humanities
Coloring in the world of others: color use in visual orientalism, 1890-1920
T. Smits and M. Wevers
Business
Poison or catalyst? How do energy saving targets matter for firm-level productivity in China
P. Zhang, A. Zhang, et al.
Engineering and Technology
A star-nose-like tactile-olfactory bionic sensing array for robust object recognition in non-visual environments
M. Liu, Y. Zhang, et al.
