🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

some vision tasks are fake, i won't tell you which ones

if your robotics startup uses jax you're ngmi

"robot, throw the sponge away from the table" *clicks submit*😈

kids are so competitive these days. they got my 2 week old niece studying black and white circles

instructions unclear, accidentally fine-tuned GR00T on cartpole

Excited to announce the 1st Workshop on Robot Hardware-Aware Intelligence @ #RSS2025 in LA! We’re bringing together interdisciplinary researchers exploring how to unify hardware design and intelligent algorithms in robotics! Full info: rss-hardware-intelligence.github.io Robotics: Science and Systems

For many empirical CS/AI research problems: If your project doesn't start with a rather lengthy engineering phase where you just get something to work as well as it can without novelty, you're likely just setting yourself up for non-additive success thanks to weak baselines.

Practical Efficiency of Muon for Pretraining "We demonstrate that Muon, the simplest instantiation of a second-order optimizer, explicitly expands the Pareto frontier over AdamW on the compute-time tradeoff. We find that Muon is more effective than AdamW in retaining data

Chris Paxton ngl all robot machine learning is about making the real world look in distribution

Over the last few months, the team at hud has made a lot of evals and environments. When we first started, we ran into a lot of problems: 1) Hosting CUA evals is annoying 2) Creating RL environments and problems is hard 3) Reviewing trajectories was super tedious 4) There

bro doesn’t have a product but boasting about internal reorg