“We believe that hardware needs to be built in-house… scaling human data is extremely important… other companies bet on scaling robot data only”
♫
Said on Latent Space: The AI Engineer Podcast
“We already committed to pre-training on humans because we knew our robots will be like human; the transfer is actually there and very helpful”
Latent Space: The AI Engineer Podcast · Dyna-2: General Purpose Physical Agents with Hour-Long Autonomy - Jason Ma, Dyna Robotics
Play from 29:14
29:14
Audio streams from the publisher’s own feed. Verbatim does not host or copy it.
What was described
pre-train robot control models on human motion data before the robot hardware is built
Said nearby on the same subject
“Our robots fold thousands of napkins per day… the model can be put in different robots across locations”
♫
Said on Latent Space: The AI Engineer Podcast
“We built this 100 robot fleet called Data Factory One in Boston. Yeah. This is to our knowledge the largest sort of semi-humanoid data collection farm in the United States.”
♫
Said on The a16z Show
“We were the, you know, the, really the first in the industry to adopt, um, robotic process automation. So we built RPA, our first RPA models probably 10, 11 years ago.”
FA
First Advantage
Not Another CEO Podcast ·
Moments like this, every Monday
The strongest new moments on your topics, in the speakers' own words. Free for three topics.
No card. No password. One-click unsubscribe.
Track a competitor?
Pro watches up to 25 names or phrases and emails you the morning after they are said.
See Pro, $29 a monthVerbatim quotes a short extract and links to the episode; the episode belongs to its publisher. Remove a show.