Twitxy Saved

Bo Dai

@daibond_alpha

Assistant Professor at Georgia Tech @GTCSE, Senior Staff Research Scientist at @GoogleDeepMind | ex @GoogleBrain
California, USA bo-dai.github.io Joined October 2012
836 Following 3.3K Followers 260 Posts

Latest posts 5

Pinned post

Bo Dai @daibond_alpha
One Kinex for any robot, making scaling across different robots possible!
Chenxiao Gao Verified account @chenxiaogao Oct 15
Introducing Kinex, the harness that connects coding agents to any physical robot, unlocking scalable learning in the physical world.

Tell Kinex about your robot; it then carries out tasks and learns from every run. We demonstrate Kinex across diverse real-world applications in a zero-shot setting, from bimanual manipulation to quadruped robots.

🧵1/5
View quoted post on X
5 replies 0 reposts 8 likes 1.6K views View on X
Bo Dai @daibond_alpha
Congrats to @fei_yang_wu and @chenxiaogao! It is such an exciting journey to make spectral representation work on real robots with astonishing results!
Feiyang Wu Verified account @fei_yang_wu Nov 13
Wonderful experience at CoRL this year! As a perfect closing gift, we won the best paper at the Compositional and Learning workshop 🤩🎉🥳 @chenxiaogao @daibond_alpha @anqiwu_angela

Check out how we steer, mix, chain humanoid motions🏃‍♂️‍➡️🤹‍♂️🕺: spectral-skill.github.io.
View quoted post on X
Bo Dai @daibond_alpha
Yes. It is. So it is essential a memory-efficient implementation of "one-step" on-policy policy gradient update w.r.t. the sampling policy, whose gradient calculation is actually divided into several batches.
Nan Jiang @nanjiang_cs Sep 19
@YouJiacheng consider linear softmax policy. the proposed correction does nothing and simply calculate the gradient under the sampling distribution.
View quoted post on X
0 replies 0 reposts 15 likes 2.2K views View on X