Skip to content

Latest commit

 

History

9 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 

Repository files navigation

RL for CUDA Kernel Optimization

Reinforcement Learning framework for automatic optimization of CUDA kernels. The project models kernel transformation as an MDP where an agent learns to apply optimizations (fusion, fission, unrolling) to maximize execution speed.

Features

  • RL-based kernel optimization
  • Custom CUDA environment with performance feedback
  • Focus on key benchmarks (MatMul, element-wise ops, reductions)

Goal

Discover optimization strategies that outperform heuristic or manual tuning.

About

No description, website, or topics provided.

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages