How to Train Your Robot
Construction Physics
Robot training is presented as a data problem tied to the scaling idea that better AI comes from bigger models, more training data, and more compute. The article notes Epoch AI’s estimate that the existing stock of quality, accessible text data is about 300 trillion tokens, while there is no comparable stock of public robot training data. As a result, it emphasizes that robotics research must rely on methods like teleoperation, reinforcement learning with robots or simulations, and simulation-to-real approaches to generate enough training data.
Why it matters
Training robot AI depends on lots of examples of something being done correctly, but generating this data is expensive. The newsletter notes researchers are trying to gather enormous amounts of robot task training data to see whether scaling trends similar to LLMs show up for robotics.