Transfer learning is a machine learning method where a model developed for a task is reused as the starting point for a model on a second task. We use a pretrained model which is trained for one task. This model can be retrained for a similar task with a smaller dataset.
Transfer Learning gives higher accuracy compared to training models from scratch
Examples:
- VGG-16 - Image Recognition model
- ResNet-50
- Inceptionv3
- MobileNet V2