hardmaru 😷: DiffTaichi: Differentiable Programming for Physical Simulation
“Using our differentiable programs, neural network controllers are typically optimized within only tens of iterations.”
When we have good priors about the world, it makes sense to use them!
9 replies, 725 likes
Lukas Mosser: More differentiable physics frameworks for all your deep learning needs!
0 replies, 10 likes
Greg: This is one of the most incredible papers I've read in the last year or so:
Differentiable Programming for Physical Simulation: https://arxiv.org/abs/1910.00935
I think you two might be interested in this!
0 replies, 3 likes
Eric Jang @ CoRL: Furthermore, the authors extend this software to implement differentiable physics simulators for a very impressive variety of sims (3d voxel mpm, liquids, rigid body) and even differentiable rendering ! (https://arxiv.org/pdf/1910.00935.pdf)
1 replies, 0 likes
Found on Oct 28 2019 at https://arxiv.org/pdf/1910.00935.pdf