This is a modal window.
Beginning of dialog window. Escape will cancel and close the window.
End of dialog window.
Note: This video may not reflect the current shipping version
Title: Scalable Adaptive Domani Randomization with BayesSimG
Author: Rika Antonova
This video introduces BayesSimIG: a library that offers large-scale simulation parameter inference with end-to-end GPU acceleration. Viewers will learn how to obtain mixture posteriors over simulation parameters using a single (or a few) real trajectories and use these to enable adaptive domain randomization for Sim-to-Real training in Isaac Gym.
Title: Operational space control for robot manipulation
Time: 14:57 to 20:27 Author: Josiah Wong
Additionally, viewers will learn about end-to-end GPU accelerated Operational Space Control in Isaac Gym and its applications to solving challenging robotic manipulation tasks.