Shift neural networks reduce computation complexity by removing expensive multiplication operations and quantizing continuous weights into low-bit discrete values, which are fast and energy efficient compared to conventional neural networks. However, existing shift networks are sensitive to the weight initialization, and also yield a degraded performance caused by vanishing gradient and weight sign freezing problem. To address these issues, we propose S low-bit re-parameterization, a novel technique for training low-bit shift networks. Our method decomposes a discrete parameter in a sign-sparse-shift 3-fold manner. In this way, it efficiently learns a low-bit network with a weight dynamics similar to full-precision networks and insensitive to weight initialization. Our proposed training method pushes the boundaries of shift neural networks and shows 3-bit shift networks out-performs their full-precision counterparts in terms of top-1 accuracy on ImageNet.
@article{arxiv.2107.03453,
title = {$S^3$: Sign-Sparse-Shift Reparametrization for Effective Training of Low-bit Shift Networks},
author = {Xinlin Li and Bang Liu and Yaoliang Yu and Wulong Liu and Chunjing Xu and Vahid Partovi Nia},
journal= {arXiv preprint arXiv:2107.03453},
year = {2021}
}