English

Loss-aware Weight Quantization of Deep Networks

Machine Learning 2018-05-11 v2

Abstract

The huge size of deep networks hinders their use in small computing devices. In this paper, we consider compressing the network by weight quantization. We extend a recently proposed loss-aware weight binarization scheme to ternarization, with possibly different scaling parameters for the positive and negative weights, and m-bit (where m > 2) quantization. Experiments on feedforward and recurrent neural networks show that the proposed scheme outperforms state-of-the-art weight quantization algorithms, and is as accurate (or even more accurate) than the full-precision network.

Keywords

Cite

@article{arxiv.1802.08635,
  title  = {Loss-aware Weight Quantization of Deep Networks},
  author = {Lu Hou and James T. Kwok},
  journal= {arXiv preprint arXiv:1802.08635},
  year   = {2018}
}
R2 v1 2026-06-23T00:31:40.828Z