“I made an RL agent that really utilized GPUs. I could scale like linear with compute… it was very efficient and easy to extend.”
♫
Said on Training Data
“To use the GPU power more efficiently, you can aggregate the API code from different users to fill in the computes.”
Ming-Yu Liu, Nvidia · Practical AI · Open models and the future of Physical AI with NVIDIA
Play from 17:22
17:22
Audio streams from the publisher’s own feed. Verbatim does not host or copy it.
What was described
Aggregate API calls from multiple users on a GPU to keep the GPU compute fully utilized for autoregressive transformer models (AOM)
Said nearby on the same subject
“You’d be surprised how underutilized GPUs are… if you also just aggregate things globally… squeeze out of the available hardware to lower these costs”
♫
Said on Unsupervised Learning with Jacob Effron
“If you're a neocloud, you can buy the GPUs, deploy them, sell a strip of GPU futures, and then just sell on-demand compute…sell six month contracts…post a big percentage upfront”
♫
Said on Odd Lots
“…hard to justify me taking up hundreds of GPUs for a substantial amount of time … unless they’re in line with the company's goals or there’s downtime and underutilized GPUs.”
EB
Eric Bigelow Goodfire
"The Cognitive Revolution" ·
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.