Transfer learning

Using pre-trained models / their trained weights as a starting point to train a model for a different data set.

Use pre-trained net, initialize last layers with random weights.

Options: Train new layers of network, or even more layers.

Prereqs:

- Much data for existing model, for new model only few data - Pre-trained model needs to generalize

Pre-trained models which generalize can be used as a starting model, e.g. https://github.com/KaimingHe/deep-residual-networks ResNet models.

For CNNs more generic features are usually contained in the earlier layers.