Open Peer Review. Open Publishing. Open Access. Open Discussion. Open Directory. Open Recommendations. Open API. Open Source.
Shake-Shake regularization of 3-branch residual networks
Feb 17, 2017 (modified: Mar 15, 2017)ICLR 2017 workshop submissionreaders: everyone
Abstract:The method introduced in this paper aims at helping computer vision practitioners faced with an overfit problem. The idea is to replace, in a 3-branch ResNet, the standard summation of residual branches by a stochastic affine combination. The largest tested model improves on the best single shot published result on CIFAR-10 by reaching 2.86% test error. Code is available at https://github.com/xgastaldi/shake-shake
TL;DR:Reduce overfit by replacing, in a 3-branch ResNet, the standard summation of residual branches by a stochastic affine combination
Keywords:Computer vision, Deep learning, Supervised Learning
Enter your feedback below and we'll get back to you as soon as possible.