ulamaca / l0_regularization Goto Github PK
View Code? Open in Web Editor NEWTensorFlow Implementation of an approximate version of L_0 regularization from Louizos et al. 2017. The implementation is generally applicable to both weights/biases and units activation in neural networks.