Majid Khadiv

Set-based training can shrink the reachable-action radius of visuomotor policies without sacrificing performance, outperforming traditional methods by a significant margin.

PAC-DP achieves remarkable improvements in robotic manipulation tasks, especially under low-data conditions, by leveraging a principled PAC-Bayes framework.

Sparse demonstrations can now effectively bootstrap humanoid loco-manipulation learning, reducing the need for constant human oversight.

By systematically guiding diffusion policies to explore underrepresented behaviors, this framework uncovers novel trajectories that traditional methods often overlook.

Automated discovery of complex humanoid manipulation skills could revolutionize how robots learn and adapt to new tasks without human input.