About this document

Regularization in Deep Learning Models by Yun Su is a document available to read on EtoBox.

- The document discusses regularization techniques for deep learning models to reduce overfitting on small training datasets. - It presents a neural network model for predicting whether a football player will hit the ball, trained on a 2D dataset from past games. Without regularization, the model overfits the training data. - L2 regularization and dropout are then introduced as methods to reduce overfitting. The model is retrained with these techniques to improve generalization to new examples.

Author

Yun Su

Language

EN