Correlated Input-Dependent Label Noise in Large-Scale Image Classification

by   Mark Collier, et al.

Large scale image classification datasets often contain noisy labels. We take a principled probabilistic approach to modelling input-dependent, also known as heteroscedastic, label noise in these datasets. We place a multivariate Normal distributed latent variable on the final hidden layer of a neural network classifier. The covariance matrix of this latent variable, models the aleatoric uncertainty due to label noise. We demonstrate that the learned covariance structure captures known sources of label noise between semantically similar and co-occurring classes. Compared to standard neural network training and other baselines, we show significantly improved accuracy on Imagenet ILSVRC 2012 79.3 new state-of-the-art result on WebVision 1.0 with 76.6 datasets range from over 1M to over 300M training examples and from 1k classes to more than 21k classes. Our method is simple to use, and we provide an implementation that is a drop-in replacement for the final fully-connected layer in a deep classifier.


page 1

page 7

page 15


Noisy Label Learning for Large-scale Medical Image Classification

The classification accuracy of deep learning models depends not only on ...

CleanNet: Transfer Learning for Scalable Image Classifier Training with Label Noise

In this paper, we study the problem of learning image classification mod...

Class2Str: End to End Latent Hierarchy Learning

Deep neural networks for image classification typically consists of a co...

Analysis of Softmax Approximation for Deep Classifiers under Input-Dependent Label Noise

Modelling uncertainty arising from input-dependent label noise is an inc...

Visual Confusion Label Tree For Image Classification

Convolution neural network models are widely used in image classificatio...

Representing Camera Response Function by a Single Latent Variable and Fully Connected Neural Network

Modelling the mapping from scene irradiance to image intensity is essent...

Self-informed neural network structure learning

We study the problem of large scale, multi-label visual recognition with...