%0 Journal Article
%T One weird trick for parallelizing convolutional neural networks
%A Alex Krizhevsky
%J Computer Science
%D 2014
%I arXiv
%X I present a new way to parallelize the training of convolutional neural networks across multiple GPUs. The method scales significantly better than all alternatives when applied to modern convolutional neural networks.
%U http://arxiv.org/abs/1404.5997v2