PPO, visualized inside Isaac Lab
Follow one training iteration from thousands of parallel robot worlds to a shuffled PPO update.
9 minute readRuben D'Sa · Field notes
Notes on robot learning, simulation, and the systems that connect them.
Follow one training iteration from thousands of parallel robot worlds to a shuffled PPO update.
9 minute readIsaac Lab supplies the world; RLinf supplies the distributed post-training runtime.
10 minute read