BriefGPT.xyz
May, 2019
深度神经网络中数据表示的内在维度
Intrinsic dimension of data representations in deep neural networks
HTML
PDF
Alessio Ansuini, Alessandro Laio, Jakob H. Macke, Davide Zoccolan
TL;DR
研究了深度神经网络的几何属性和数据表示的内在维度,发现最后一个隐藏层的内在维度预测测试集合的分类准确性,这证明了可以广泛应用的神经网络是将数据转换为低维非线性流形的网络。
Abstract
deep neural networks
progressively transform their inputs across multiple processing layers. What are the geometrical properties of the representations learned by these networks? Here we study the
intrinsic dimensionali
→