What does “Transfer Learning” mean?
Applying knowledge learned in one task or domain to improve another task.
How Transfer Learning Works
Learn a general skill on a large task
Reuse the learned weights
Train briefly on your small dataset
Strong result with little data.
Practical Example
A vision model trained on millions of everyday photos already knows edges and shapes. A clinic reuses it and trains it on a few hundred skin images, reaching useful accuracy without starting from scratch.
Why It Matters
It makes AI practical for small projects, because most of the learning is already done.
Common Questions About Transfer Learning
Q: Transfer learning vs fine-tuning?
Fine-tuning is a common form of transfer learning where the pre-trained model's weights are updated.
Q: When does it fail?
If the new task is very different from the original, the transferred knowledge may not help.
Q: Does it need less data?
Usually much less than training from scratch.