Skip to main content

Taichi

DiffTaichi: Differentiable Programming for Physical Simulation
 Paper arXiv Paper arXiv Code



Differentiable programming in Taichi allows you to optimize neural network controllers efficiently with brute-force gradient descent, instead of using reinforcement learning. It is much more efficient than REINFORCE update. The DiffTaichi differentiable programming framework is now officially part of Taichi.

Official Materials​