On robotics
- Learning like a baby: a plan for an embodied world model
Google DeepMind's Gemini Robotics ER 2 gives robots a better high-level brain. The part I still think nobody has built is the body-first learning underneath. Here's the research plan I'd run.
3 min reads likes comments - 62 hours of robot data
Meta's V-JEPA 2 learns a world model from a million hours of video, then learns to plan robot actions from 62 hours of robot data. Promising, and still missing touch.
2 min reads likes comments - Helix has a fast brain and a slow brain. So does my golf swing
Figure's Helix splits robot control into a slow model that understands and a fast one that moves. A golf swing works the same way, which is why thinking during it ruins it.
2 min reads likes comments - Mobile ALOHA cooks shrimp for $32,000
Stanford's two-armed robot on a wheeled base learned to cook, wipe spills and call an elevator from about fifty demonstrations per task. Cheap hardware plus teleoperation is changing robot data.
2 min reads likes comments - Optimus sorts blocks. A toddler falls down a thousand times
Tesla's new video shows Optimus sorting colored blocks with a neural network trained end to end. Curated robot demos and messy child learning, and which one scales.
2 min reads likes comments - Optimus walked on stage. Watch its hands
A year after the dancer in the suit, Tesla showed a real humanoid prototype. It walked slowly and waved. The legs get the attention. The hands are the harder and more important problem.
2 min reads likes comments - One network, 604 tasks
DeepMind's Gato plays Atari, captions images, chats and stacks blocks with a real robot arm, all with the same weights. Turning actions into tokens is the interesting part.
2 min reads likes comments - A dancer in a spandex suit, and why the humanoid form matters
Tesla's AI Day announced a humanoid robot and then showed a person dancing in a robot costume. The demo was a joke. The idea is right.
2 min reads likes comments - A robot hand with a hundred years of practice
OpenAI's Dactyl learned to rotate a block with a human-like robot hand, entirely in simulation. Why hands, and why randomizing the simulator is the clever part.
2 min reads likes comments - Nobody taught it to walk
DeepMind's locomotion agents learned to run, jump and duck from a reward for moving forward and a varied course. The flailing arms are the point.
2 min reads likes comments - The long tail of driving
Uber will put self-driving cars on Pittsburgh streets next month, six weeks after the first Autopilot fatality became public. The hard part of driving is the rare case.
2 min reads likes comments - The best robotics video ever made involves a hockey stick
Boston Dynamics' new Atlas gets shoved, loses its box and gets knocked flat, and keeps going. Why recovering is harder than walking.
2 min reads likes comments