Unsupervised Deep Learning by Neighbourhood Discovery
This addresses the scalability and deployment limitations of supervised CNNs in computer vision by eliminating the need for labeled data, though it appears incremental as it builds on existing unsupervised learning paradigms.
The authors tackled the problem of training deep convolutional neural networks without manual labels by introducing an unsupervised approach that discovers sample neighborhoods to learn class decision boundaries iteratively. Their method outperformed state-of-the-art unsupervised models on six image classification benchmarks, including coarse-grained and fine-grained object categorization.
Deep convolutional neural networks (CNNs) have demonstrated remarkable success in computer vision by supervisedly learning strong visual feature representations. However, training CNNs relies heavily on the availability of exhaustive training data annotations, limiting significantly their deployment and scalability in many application scenarios. In this work, we introduce a generic unsupervised deep learning approach to training deep models without the need for any manual label supervision. Specifically, we progressively discover sample anchored/centred neighbourhoods to reason and learn the underlying class decision boundaries iteratively and accumulatively. Every single neighbourhood is specially formulated so that all the member samples can share the same unseen class labels at high probability for facilitating the extraction of class discriminative feature representations during training. Experiments on image classification show the performance advantages of the proposed method over the state-of-the-art unsupervised learning models on six benchmarks including both coarse-grained and fine-grained object image categorisation.