Reinforcement learning has transformed games and robotics, but fluid dynamics lacks shared benchmarks and standardized environments. Every controller is tuned for a single geometry and operating condition — progress cannot accumulate, transfer, or be compared. The field needs its
Fluid Dynamics Needs Shared Benchmarks for Reinforcement Learning Controllers
By
–
