Open Peer Review. Open Publishing. Open Access. Open Discussion. Open Directory. Open Recommendations. Open API. Open Source.
Regularizing Neural Networks by Penalizing Confident Output Distributions
Gabriel Pereyra, George Tucker, Jan Chorowski, Lukasz Kaiser, Geoffrey Hinton
Nov 04, 2016 (modified: Jan 21, 2017)ICLR 2017 conference submissionreaders: everyone
Abstract:We propose regularizing neural networks by penalizing low entropy output distributions. We show that penalizing low entropy output distributions, which has been shown to improve exploration in reinforcement learning, acts as a strong regularizer in supervised learning. We connect our confidence penalty to label smoothing through the direction of the KL divergence between networks output distribution and the uniform distribution. We exhaustively evaluate our proposed confidence penalty and label smoothing (uniform and unigram) on 6 common benchmarks: image classification (MNIST and Cifar-10), language modeling (Penn Treebank), machine translation (WMT'14 English-to-German), and speech recognition (TIMIT and WSJ). We find that both label smoothing and our confidence penalty improve state-of-the-art models across benchmarks without modifying existing hyper-parameters.
TL;DR:We show that penalizing low entropy output distributions, which has been shown to improve exploration in reinforcement learning, acts as a strong regularizer in supervised learning.