BriefGPT.xyz
Jun, 2014
使用有噪声的标签训练卷积神经网络
Learning from Noisy Labels with Deep Neural Networks
HTML
PDF
Sainbayar Sukhbaatar, Rob Fergus
TL;DR
本论文探究了在实际情况中,数据标注不可避免地有可能存在噪声标签的情况下,利用引入噪声层的方法,在Convolutional Network模型上进行判别性训练以获得更好的性能。同时在多个数据集上进行了实验,包括在ImageNet分类基准测试上的大规模试验。
Abstract
We propose several simple approaches to training deep neural networks on data with
noisy labels
. We introduce an extra
noise layer
into the network which adapts the network outputs to match the noisy label distri
→