arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2010.04871cs.CV
收藏

通过带噪声监督的学习训练二元神经网络

Training Binary Neural Networks through Learning with Noisy Supervision

Institute of Software, CAS(中国科学院软件研究所);University of Chinese Academy of Sciences(中国科学院大学);Noah’s Ark Lab, Huawei Technologies(华为技术有限公司诺亚方舟实验室);University of Macau(澳门大学);University of Sydney(悉尼大学)

Kai Han, Yunhe Wang, Yixing Xu, Chunjing Xu, Enhua Wu, Chang Xu

摘要:

本文从学习的角度对神经网络上的二值化操作进行了形式化。与经典的手工制作的规则(\eg硬阈值)二进制化全精度神经元相比,我们建议学习从全精度神经元到目标二进制神经元的映射。每个单独的权重条目将不会被单独二进制化。相反,它们被作为一个整体来完成二值化,就像它们在生成卷积特征时一起工作一样。为了帮助二值化映射的训练,经过符号运算的全精度神经元被视为一些辅助监督信号,这是噪音,但仍然有价值的指导。因此,引入无偏估计器来减轻监督噪声的影响。基准数据集上的实验结果表明,所提出的二值化技术取得了一致的改善基线。

英文摘要:

This paper formalizes the binarization operations over neural networks from a learning perspective. In contrast to classical hand crafted rules (\eg hard thresholding) to binarize full-precision neurons, we propose to learn a mapping from full-precision neurons to the target binary ones. Each individual weight entry will not be binarized independently. Instead, they are taken as a whole to accomplish the binarization, just as they work together in generating convolution features. To help the training of the binarization mapping, the full-precision neurons after taking sign operations is regarded as some auxiliary supervision signal, which is noisy but still has valuable guidance. An unbiased estimator is therefore introduced to mitigate the influence of the supervision noise. Experimental results on benchmark datasets indicate that the proposed binarization technique attains consistent improvements over baselines.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑