Training Data · David Silver and Lasse Espeholt of Ineffable Intelligence: A Superlearner for the Age of Experience
Audio streams from the publisher’s own feed. Verbatim does not host or copy it.
Created “Seed RL”, a system that runs on racks of TPUs for both inference and training across flexible environments
“… very good RL controllers that are probably trained from motion imitation, and then fine-tuned with good RL that suits the robot's body… for the specific task of running really fast.”
“They're applying these models to the RL environments themselves… hack this environment… fixing those… lower rates of cheating downstream”
“We have a phrase we call the Moore's Law of RL environments, which basically means, like, every six to eight months, we've noticed that the complexity of the environment… roughly doubles.”
“They have to devote the time… have peers that have learned how to train and mentor others… create a non-judgmental environment.”
The strongest new moments on your topics, in the speakers' own words. Free for three topics.
No card. No password. One-click unsubscribe.
Pro watches up to 25 names or phrases and emails you the morning after they are said.
See Pro, $29 a monthVerbatim quotes a short extract and links to the episode; the episode belongs to its publisher. Remove a show.