arXiv Analytics

Sign in

arXiv:1604.03058 [cs.CV]AbstractReferencesReviewsResources

High Performance Binarized Neural Networks trained on the ImageNet Classification Task

Xundong Wu

Published 2016-04-11Version 1

We trained Binarized Neural Networks (BNNs) on the high resolution ImageNet LSVRC-2102 dataset classification task and achieved a good performance. With a moderate size network of 10 layers, we obtained top-5 classification accuracy rate of 81 percent on validation set which is much better than previous published results. We expect training networks of a much better performance through increase network depth would be straight forward by following our current strategies. A detailed discussion on strategies used in the network training is included as well as preliminary analysis.

Related articles: Most relevant | Search more
arXiv:1911.09074 [cs.CV] (Published 2019-11-20)
Search to Distill: Pearls are Everywhere but not the Eyes
arXiv:1608.08614 [cs.CV] (Published 2016-08-30)
What makes ImageNet good for transfer learning?
arXiv:2007.15036 [cs.CV] (Published 2020-07-29)
Generative Classifiers as a Basis for Trustworthy Computer Vision