What initially looks like playing with hyperparameters brings new life to a somewhat older approach. From Alex's tweet:
FreezeOut: it's like layerwise pretraining for #DeepLearning hipsters who weren't around before 2009 ;-) https://t.co/q3hPVvXaa6 pic.twitter.com/t3fshovQhe— Alex J. Champandard (@alexjc) 16 juin 2017
FreezeOut: Accelerate Training by Progressively Freezing Layers by Andrew Brock, Theodore Lim, J.M. Ritchie, Nick Weston
The early layers of a deep neural net have the fewest parameters, but take up the most computation. In this extended abstract, we propose to only train the hidden layers for a set portion of the training run, freezing them out one-by-one and excluding them from the backward pass. We empirically demonstrate that FreezeOut yields savings of up to 20% wall-clock time during training with 3% loss in accuracy for DenseNets on CIFAR.
Join the CompressiveSensing subreddit or the Google+ Community or the Facebook page and post there !
Liked this entry ? subscribe to Nuit Blanche's feed, there's more where that came from. You can also subscribe to Nuit Blanche by Email, explore the Big Picture in Compressive Sensing or the Matrix Factorization Jungle and join the conversations on compressive sensing, advanced matrix factorization and calibration issues on Linkedin.
No comments:
Post a Comment