"Foundation" models (Practical AI #172)
Changelog Master Feed
00:00
Using Foundation Models to Build Scalable Data Sets?
The idea is we're really talking about scale, and you can't build large data sets if you're always thinking just in terms of supervised learning. Not all researchers and practitioners can create their own foundational model because I don't have like racks of GPUs or the computational resources to actually create a model on this scale. If there's an unexpected behavior or bias in that foundation model, then that actually filters through down to a huge number of downstream tasks and applications.
Transcript
Play full episode
