The paper's directional selection step is a genetic algorithm, but is only activated only when there is sufficient performance difference. Likewise, evolutionary theory states that selection can only act on existing variation within a population: if all agents perform

Thank you for all your help throughout the paper! It all started with SAPG. I look forward to seeing more work on improving the scalability and efficiency of RL.

Tired of tuning PPO or blaming it on reward, task design, etc.? Introducing EPO -- our second (and hopefully final :) attempt at fixing PPO at scale! Contrary to intuition, as the batch size or data increases, PPO saturates due to a lack of diversity in sampling. We proposed a

Tactile sensing is gaining traction, but slowly. Why? Because integration remains difficult. But what if adding touch sensors to your robot was as easy as hitting “print”? Introducing eFlesh: a 3D-printable, customizable tactile sensor. Shape it. Size it. Print it. 🧶👇

Great to see it launched! Congrats to Vikash Kumar and team!

Three years ago, when we began exploring learning from video—most tasks were just pick-and-place. With PSAG, we enabled one-shot learning of deformable object manipulation from YouTube. Now, this paper pushes it further, tackling a wider range of tasks via visual FMs w.o. demos!

Stay tuned 👀

Thrilled to be part of this incredible journey with such an amazing team working toward AGI. Excited for what’s to come!

In nature, animals learn to perceive their environment before mastering complex locomotion. Skild AI’s robot brain confirms this principle: with Pix2Loco, we achieved arguably the most robust locomotion skill to date. What comes next👀?

This is a strong case for any robot, one brain.I once held that such generalizability did not exist in nature—until they reminded me of the caterpillar. Every day, I feel proud of what we are building here. To the god of AGI: I bow to my robot each day, so the person isn’t me😜.