Spatial contrasting for deep unsupervised learning
This addresses the challenge of leveraging unlabeled data for visual tasks, offering a method that complements supervised learning without requiring specialized architectures.
The paper tackles the problem of unsupervised training for convolutional networks by introducing a spatial contrasting approach, which can be integrated into conventional architectures and trained with standard techniques like SGD and back-propagation.
Convolutional networks have marked their place over the last few years as the best performing model for various visual tasks. They are, however, most suited for supervised learning from large amounts of labeled data. Previous attempts have been made to use unlabeled data to improve model performance by applying unsupervised techniques. These attempts require different architectures and training methods. In this work we present a novel approach for unsupervised training of Convolutional networks that is based on contrasting between spatial regions within images. This criterion can be employed within conventional neural networks and trained using standard techniques such as SGD and back-propagation, thus complementing supervised methods.