DiffTaichi: Differentiable Programming for Physical Simulation

Yuanming Hu, Luke Anderson, Tzu-Mao Li, Qi Sun, Nathan Carr, Jonathan Ragan-Kelley, Fredo Durand

Keywords: nlp, optimal control, optimization, robotics

Thurs Session 4 (17:00-19:00 GMT) [Live QA] [Cal]
Thurs Session 5 (20:00-22:00 GMT) [Live QA] [Cal]

Abstract: We present DiffTaichi, a new differentiable programming language tailored for building high-performance differentiable physical simulators. Based on an imperative programming language, DiffTaichi generates gradients of simulation steps using source code transformations that preserve arithmetic intensity and parallelism. A light-weight tape is used to record the whole simulation program structure and replay the gradient kernels in a reversed order, for end-to-end backpropagation. We demonstrate the performance and productivity of our language in gradient-based learning and optimization tasks on 10 different physical simulators. For example, a differentiable elastic object simulator written in our language is 4.2x shorter than the hand-engineered CUDA version yet runs as fast, and is 188x faster than the TensorFlow implementation. Using our differentiable programs, neural network controllers are typically optimized within only tens of iterations.

Similar Papers

Learning to Control PDEs with Differentiable Physics
Philipp Holl, Nils Thuerey, Vladlen Koltun,
Infinite-Horizon Differentiable Model Predictive Control
Sebastian East, Marco Gallieri, Jonathan Masci, Jan Koutnik, Mark Cannon,
Lagrangian Fluid Simulation with Continuous Convolutions
Benjamin Ummenhofer, Lukas Prantl, Nils Thuerey, Vladlen Koltun,