Google Brain uncovers representation structure differences between CNNs and vision transformers. Andrey Kuznetsov: I found this pretty interesting. Some of the results are not very surprising, as you said. So for instance, in the early layers, closer to input, there's more global information and then resnets,. which makes sense due to the design transformers.
Our 70th episode with a summary and discussion of last week's big AI news!
This week:
- Applications
- Research
- Ethics
- Fun
Subscribe: RSS | iTunes | Spotify | YouTube
Find this and more in our text version of this news roundup: https://lastweekin.ai/p/131
Music: Deliberate Thought, Inspired by Kevin MacLeod (incompetech.com)