Deep Convolutional Transform Learning

• Published in: Neural Information Processing Vol. 1333; pp. 300 - 307
• Main Authors: Maggu, Jyoti, Majumdar, Angshul, Chouzenoux, Emilie, Chierchia, Giovanni
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2020; Springer International Publishing
• Series: Communications in Computer and Information Science
• Subjects: Alternating minimization; Classification; Clustering; Convolutional neural networks; Deep learning; Proximal methods; Transform learning
• ISBN: 9783030638221; 3030638227
• ISSN: 1865-0929; 1865-0937
• DOI: 10.1007/978-3-030-63823-8_35

This work introduces a new unsupervised representation learning technique called Deep Convolutional Transform Learning (DCTL). By stacking convolutional transforms, our approach is able to learn a set of independent kernels at different layers. The features extracted in an unsupervised manner can then be used to perform machine learning tasks, such as classification and clustering. The learning technique relies on a well-sounded alternating proximal minimization scheme with established convergence guarantees. Our experimental results show that the proposed DCTL technique outperforms its shallow version CTL, on several benchmark datasets.