🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Can neuro-inspired ANN architectures be useful for motor control in quadruped robots? We translate neural circuits in the limbs and spinal cord of mammals into an ANN architecture controlling quadruped locomotion. w/ Venkatesh, Lerrel Pinto, @neurograce

Wouldn't it be great if we could train robots without any teleoperation! In our latest paper, we train robots to mimic a human video of the task by simply matching the object features using RL. We only need one video and under an hour of robot training.

If you want robots that can just live with you & help 24/7, it needs to build & update its memory on the fly. Current semantic memory representations like VoxelMap from OK-Robot can't change with the world. That's why we built DynaMem: dynamic memory for a changing, open world!

Excited to share that I'll be joining University of California at Irvine as a CS faculty in '25!🌟 Faculty apps: Krishna Murthy, Zhuang Liu & I share our tips: unnat.github.io/notes/Hidden_C… PhD apps: I'm looking for students in vision, robot learning, & AI4Science. Details👇

Thrilled to share our work on Navigation World Models! It’s exciting to see how world models empower planning for navigation and shed light on agents with reasoning abilities. Such a fun collaboration with everyone! arXiv: arxiv.org/abs/2412.03572 Website: amirbar.net/nwm

New work: DINO-World Model. Planning with JEPA. ⬇️⬇️⬇️

Exciting to see Kensuke Nakamura's new work on generalizing robot safety building upon DINO-WM for real-world experiments! Congrats!

Navigation World Models was accepted to #CVPR2025 🎉 Congrats to co-authors Gaoyue Zhou, Danny Tran, trevordarrell and Yann LeCun See you in Nashville!

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.

1.5 yrs ago, we set out to answer a seemingly simple question: what are we *actually* getting out of RL in fine-tuning? I'm thrilled to share a pearl we found on the deepest dive of my PhD: the value of RL in RLHF seems to come from *generation-verification gaps*. Get ready to🤿!

Despite great advances in learning dexterity, hardware remains a major bottleneck. Most dexterous hands are either bulky, weak or expensive. I’m thrilled to present the RUKA Hand — a powerful, accessible research tool for dexterous manipulation that overcomes these limitations!

Presenting DemoDiffusion: An extremely simple approach enabling a pre-trained 'generalist' diffusion policy to follow a human-demonstration for a novel task during inference One-shot human imitation *without* requiring any paired human-robot data or online RL 🙂 1/n

What would a World Model look like if we start from a real embodied agent acting in the real world? It has to have: 1) A real, physically grounded and complex action space—not just abstract control signals. 2) Diverse, real-life scenarios and activities. Or in short: It has to

Generalization needs data. But data collection is hard for precise tasks like plugging USBs, swiping cards, inserting plugs, and keying locks. Introducing robust, precise VisuoTactile Local (ViTaL) policies: >90% success rates from just 30 demos and 45 min of real-world RL.🧶⬇️