Siddhant Haldar
@haldar_siddhant • 1,032 subscribers
Excited about generalizing AI | PhD student @CILVRatNYU | Undergrad @IITKgp
Videos
Current robot policies often face a tradeoff: they're either precise (but brittle) or generalizable (but imprecise). We present ViTaL, a framework that lets robots generalize precise, contact-rich manipulation skills across unseen environments with millimeter-level precision. 🧵
Siddhant Haldar145,945 Aufrufe • vor 11 Monaten
The most frustrating part of imitation learning is collecting huge amounts of teleop data. But why teleop robots when robots can learn by watching us? Introducing Point Policy, a novel framework that enables robots to learn from human videos without any teleop, sim2real, or RL.
Siddhant Haldar69,031 Aufrufe • vor 1 Jahr
Robot foundation models are limited by costly real data, while simulation data is plentiful but visually mismatched to reality. We present Point Bridge, a method that enables zero-shot sim-to-real transfer for robot learning with minimal visual alignment.
Siddhant Haldar19,714 Aufrufe • vor 3 Monaten
Keine weiteren Inhalte verfügbar