Training & Learning
Transfer Learning
Reusing what a model learned on one task as a head start for a different but related task.
Transfer learning takes the knowledge baked into a model trained for one problem and applies it to another, instead of starting over. It is why a model pretrained on general text or images can be adapted to a specialized job with relatively little new data. The whole foundation-model approach is transfer learning at scale: train once broadly, then transfer everywhere.
Reference: Wikipedia: Transfer Learning
Related terms
Fine-Tuning
Taking a pretrained model and training it further on a smaller, focused dataset to specialize it for a task or style.
Pretraining
The first, largest training phase where a model learns general patterns from a huge amount of broad data.
Foundation Model
A large, general-purpose model trained on broad data that many different applications can be built on top of.