Main Concept
Transfer Learning (TL) is broader concept that refers to the idea to take a knowledge on one task and apply to another.
It’s a machine learning (ML) technique where a model pre-trained on one task is fine-tuned for a new related task while maintaining general capabilities
Key Points
-
The process of transfer learning involves three main steps:
-
Select a pre-trained model
-
Modify the model
-
Fine-tune the model
NOTE
Fine-tuning is used to achieve transfer learning
-
-
Domain Adaptation is a specific type of transfer learning
Example
For example, if a machine learning model can identify images of dogs, it can be trained to identify cats using a smaller image set that highlights the feature differences between dogs and cats.