Do Differentiable Simulators Give Better Policy Gradients?

H. J. Terry Suh Max Simchowitz Kaiqing Zhang Russ Tedrake

Published in: CoRR (2022)

Keyphrases

objective function
policy making
neural network
real world
website
image sequences
multiscale
pairwise
image gradient
asymptotically optimal
gradient domain
inverse reinforcement learning