What is Transfer Learning?
TL;DR
A machine learning technique that applies knowledge learned from one task to another. Dramatically improves AI development efficiency.
Transfer Learning: Definition & Explanation
Transfer learning is a machine learning technique that reuses a model trained on one task or domain for a different task or domain. For example, an image recognition model trained on millions of general images can be fine-tuned with a small set of medical images to build a highly accurate medical imaging model. The key advantage is achieving high performance with less data and shorter training time compared to training from scratch. The relationship between LLM pre-training and fine-tuning is a form of transfer learning, where foundation models like GPT and BERT are adapted to various downstream tasks. It is an indispensable technique in modern AI development.