arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2004.14214cs.LGstat.ML
收藏

量化网络中的批量归一化

Batch Normalization in Quantized Networks

Huawei Noah’s Ark Lab(华为诺亚方舟实验室)

Eyyüb Sari, Vahid Partovi Nia

摘要:

在计算硬件上实现量化的神经网络导致相当大的速度提高和存储器节省。然而,量化的深度网络很难训练,而批量归一化(BatchNorm)层在训练全精度和量化的网络中起着重要作用。大多数关于BatchNorm的研究都集中在全精度网络上,很少有研究了解BatchNorm在量化训练中的影响。我们表明BatchNorm避免了梯度爆炸,这是违反直觉的,最近在其他研究人员的数值实验中观察到的。

英文摘要:

Implementation of quantized neural networks on computing hardware leads to considerable speed up and memory saving. However, quantized deep networks are difficult to train and batch~normalization (BatchNorm) layer plays an important role in training full-precision and quantized networks. Most studies on BatchNorm are focused on full-precision networks, and there is little research in understanding BatchNorm affect in quantized training which we address here. We show BatchNorm avoids gradient explosion which is counter-intuitive and recently observed in numerical experiments by other researchers.

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。

↑