Mutual Exclusivity Loss for Semi-Supervised Deep Learning

TL;DR

This paper introduces a mutual exclusivity loss as an unsupervised regularization for semi-supervised deep learning, guiding classifiers to better utilize unlabeled data and improve object recognition accuracy.

Contribution

It proposes a novel mutual exclusivity loss that enhances semi-supervised learning by explicitly enforcing class prediction exclusivity, applicable to any backpropagation-based ConvNet training.

Findings

Improves object recognition accuracy with unlabeled data

Effective regularization for semi-supervised deep learning

Applicable to various ConvNet architectures

Abstract

In this paper we consider the problem of semi-supervised learning with deep Convolutional Neural Networks (ConvNets). Semi-supervised learning is motivated on the observation that unlabeled data is cheap and can be used to improve the accuracy of classifiers. In this paper we propose an unsupervised regularization term that explicitly forces the classifier's prediction for multiple classes to be mutually-exclusive and effectively guides the decision boundary to lie on the low density space between the manifolds corresponding to different classes of data. Our proposed approach is general and can be used with any backpropagation-based learning method. We show through different experiments that our method can improve the object recognition performance of ConvNets using unlabeled data.