Michael Cho

Michael Cho

410 Twitter followers

Follow

Plane crash survivor & huge fan of Arduino, steam engines, deep learning & Liverpool FC! Building @frodobots

Posts

My kiddo: "Cardboard is the basic unit of life!" Also, my kiddo: "This project is only 0.1% done!" Am hoping cosplay will be a legit career choice in post AGI world 🤣

Full episode dropping soon! Geeking out with @YejinKim4 @omarrayyann Max Argus on MolmoSpaces: A Large-Scale Open Ecosystem for Robot Navigation and Manipulation github.com/allenai/molmospaces… Co-hosted by @micoolcho @DJiafei

Full episode dropping soon! Geeking out with @SeonghyeonYe on DreamZero: World Action Models Are Zero-Shot Policies dreamzero0.github.io @NVIDIARobotics Co-hosted by @micoolcho @chris_j_paxton

Just a matter of time for the DreamZeroDojoCombo paper drop....can't wait!

We always need more (not less) humans-in-the-loop instead of trying to remove humans prematurely, unless we r damage sure we are really at level 5 autonomy with 6-sigma type success rate.

New essay on robot learning from human data. I like @karpathy’s idea that LLMs are “ghosts” distilled from human knowledge. In robotics, we are attempting something similar: to summon a sensorimotor ghost. Our current ritual is teleoperation. It produces data, but strips away x.com/danfei_xu/stat…

Full episode dropping soon! Geeking out with @ziwenzhuang_leo @ShaotingZ38103 on Project Instinct project-instinct.github.io Co-hosted by @micoolcho @DJiafei

Full episode dropping soon! Geeking out with @ziwenzhuang_leo @ShaotingZ38103 on Project Instinct project-instinct.github.io Co-hosted by @micoolcho @DJiafei

Full episode dropping soon! Geeking out with @Hongyu_Lii @jiahuifu_carol on NovaFlow: Zero-Shot Manipulation via Actionable Flow from Generated Videos novaflow.lhy.xyz Co-hosted by @micoolcho @chris_j_paxton

Money begets Money (hence the yield curve). Now, AI begets AI (recursively self-improving agents). The "AI Yield Curve" is steepening very quickly! A thread 🧵 - 1/n

Zero-shot generalization across infinite environments & tasks? This is the promise of world models that 1X is betting on. Tks @itsdanielho for the sharing!

Every home is different. That means that to build a useful home robot, we must be able to perform zero-shot generalization on a wide range of tasks. Humanoid company @1x_tech has a solution: world models. 1X Director of Evaluations @itsdanielho joins us on RoboPapers to talk

Loading..