Robust supervised classification with mixture models: Learning from data with uncertain labels

142 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Learning kernel logistic regression in the presence of class label noise (2014, background): “The issue has also been briefly discussed in [24,6].”
Label Noise-Tolerant Hidden Markov Models for Segmentation: Application to ECGs (2011, methodology): “also proposes a distinct robust mixture discriminant analysis [3], which consists in two steps: (i) learning an unsupervised Gaussian mixture model and (ii) computing the probability that each cluster belongs to a given class.”
Iterative Learning with Open-set Noisy Labels (2018, background): “Considering that samples from the same class should have similar high-level representations but samples mislabeled into the class do not [2], we detect noisy labels based on the representations of the pre-softmax layer.”
Multi-class classification in the presence of labelling errors (2011, background): “Often, such assumption does not hold, though (see [2, 3, 4, 5] for examples).”
Robust binary hypothesis testing under contaminated likelihoods (2014, methodology): “Somewhat more related is the mixture modeling approach of [5], [6], which attempts to learn the contamination model using the EM algorithm.”
Classification in the Presence of Label Noise: A Survey (2014, methodology): “In other words, clusters in instances can be used to build classiﬁers, what is done in [136].”