Learning Channel-Wise Interactions for Binary Convolutional Neural Networks

Ziwei Wang, Jiwen Lu, Chenxin Tao, Jie Zhou, Qi Tian; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 568-577

Abstract


In this paper, we propose a channel-wise interaction based binary convolutional neural network learning method (CI-BCNN) for efficient inference. Conventional methods apply xnor and bitcount operations in binary convolution with notable quantization error, which usually obtains inconsistent signs in binary feature maps compared with their full-precision counterpart and leads to significant information loss. In contrast, our CI-BCNN mines the channel-wise interactions, through which prior knowledge is provided to alleviate inconsistency of signs in binary feature maps and preserves the information of input samples during inference. Specifically, we mine the channel-wise interactions by a reinforcement learning model, and impose channel-wise priors on the intermediate feature maps through the interacted bitcount function. Extensive experiments on the CIFAR-10 and ImageNet datasets show that our method outperforms the state-of-the-art binary convolutional neural networks with less computational and storage cost.

Related Material


[pdf] [supp]
[bibtex]
@InProceedings{Wang_2019_CVPR,
author = {Wang, Ziwei and Lu, Jiwen and Tao, Chenxin and Zhou, Jie and Tian, Qi},
title = {Learning Channel-Wise Interactions for Binary Convolutional Neural Networks},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}