Asked about Starting from someone else's training run

When would transfer learning not help?

An answer that works

When your data is far from the pretraining distribution, such as medical scans or spectrograms against a model trained on everyday photographs. The test is empirical: if the frozen version performs near chance, the features are not transferring.

Reading an answer is not being able to give one. The section that teaches this is Starting from someone else's training run, 6 hours of Transfer learning.

Learn this properly

Also asked about Transfer learning

Why does transfer learning work?Why use a lower learning rate when fine-tuning?You have 2,000 labelled images for a new classification task. What is your approach?What is a per-layer learning rate, and when is it worth the trouble?