What Makes Transfer Learning Work For Medical Images: Feature Reuse & Other Factors — mentioned by godelski

Mon, 03 Oct 2022 19:37:35 GMT

> The early layers of the network detecting simple features like lines, curves, textures, shapes, colors, etc This is pretty common, especially with convolutions, but is not guaranteed. How you embed matters a lot. For example, there are transformers that use early convolutions for the embeddings and that makes them just work. Though too many and they are less performant (ViT tried pre-resnets which wasn't great). [0] also investigates transfer learning in medical domains and shows that CNNs depends more on statistics reuse and transformers depend more on feature reuse. > If you've seen a…

Paper and discussion history. A mention is not an endorsement.

What Makes Transfer Learning Work For Medical Images: Feature Reuse & Other Factors · HackerPapers discussions

What Makes Transfer Learning Work For Medical Images: Feature Reuse & Other Factors — mentioned by godelski