NeurIPS 2015

Learning both Weights and Connections for Efficient Neural Networks

NeurIPS 2015 songhan/Deep-Compression-AlexNet

On the ImageNet dataset, our method reduced the number of parameters of AlexNet by a factor of 9x, from 61 million to 6. 7 million, without incurring accuracy loss.

Recursive Training of 2D-3D Convolutional Networks for Neuronal Boundary Detection

NeurIPS 2015 seung-lab/znn-release

Efforts to automate the reconstruction of neural circuits from 3D electron microscopic (EM) brain images are critical for the field of connectomics.