In this paper, we introduce a new channel pruning method to accelerate very
deep convolutional neural networks.Given a trained CNN model, we propose an
iterative two-step algorithm to effectively prune each layer, by a LASSO
regression based channel selection and least square reconstruction. We further
generalize this algorithm to multi-layer and multi-branch cases. Our method
reduces the accumulated error and enhance the compatibility with various
architectures. Our pruned VGG-16 achieves the state-of-the-art results by 5x
speed-up along with only 0.3% increase of error. More importantly, our method
is able to accelerate modern networks like ResNet, Xception and suffers only
1.4%, 1.0% accuracy loss under 2x speed-up respectively, which is significant.
Code has been made publicly available.

本文提出了一种新的通道剪枝方法，能够加速非常深的卷积神经网络，通过基于 LASSO 回归的通道选择和最小二乘重构的迭代两步算法有效地修剪每一层，进一步推广到多层和多分支情况，剪枝后的 VGG-16 实现了 5 倍速度提升和仅 0.3％误差增加的最新成果，更重要的是，本方法能够加速现代网络，例如 ResNet，Xception，分别在 2 倍加速下只有 1.4％，1.0％的准确度损失，代码已公开出售。