Majid Khadiv

A novel policy optimization method bypasses critic action gradients entirely, achieving faster, more robust updates and improved exploration in continuous control.

Adaptive-MHE is presented an online sampling-based Sys-ID framework, based on moving horizon estimation (MHE), that estimates the physical parameters of objects and terrain in the environment and couples this estimate with a sampling-based model predictive controller, enabling adaptive loco-manipulation in changing and uncertain environments.

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.

The integration of LLMs into robot motion planning leads to a dramatic improvement in the efficiency of generating feasible trajectories for complex loco-manipulation tasks.