BriefGPT.xyz
Apr, 2018
深度神经网络分类的负对数似然比损失函数
Negative Log Likelihood Ratio Loss for Deep Neural Network Classification
HTML
PDF
Donglai Zhu, Hengshuai Yao, Bei Jiang, Peng Yu
TL;DR
本文提出了一种判别损失函数，与生成式训练标准不同，该函数使用正确和竞争类之间的负对数似然比，极大地提高了在CIFAR-10图像分类任务中的性能。
Abstract
In
deep neural network
, the
cross-entropy loss
function is commonly used for classification. Minimizing cross-entropy is equivalent to maximizing likelihood under assumptions of uniform feature and class distribu
→