D’après TensorFlow :
To recap: here are the most common ways to prevent overfitting in neural networks:
- Get more training data.
- Reduce the capacity of the network.
- Add weight regularization.
- Add dropout.
Two important approaches not covered in this guide are:
- data-augmentation
- batch normalization
Remember that each method can help on its own, but often combining them can be even more effective.
TensorFlow a sur GCP un notebook intéressant qui démontre l’intérêt du weight decay (L2) et du dropout sur le jeu de données Higgs.