DeepLearning1.6, Regularization, Training Base and Validation Base

views comments

The philosophy of regularization and contol of 'flexibility' is discussed and a few easy-to-implement regularization methods are discussed.