M.S. Final Oral Exam: Ramakrishnan Sundareswaran

Ramakrishnan Sundareswaran
Monday, January 17, 2022 - 2:00pm
Event Type: 

Yet Another Image Clustering Framework Using Deep Learning

Unsupervised disentangled representation learning is a long-standing problem in computer vision. This thesis aims to tackle this problem and proposes a deep learning framework for performing image clustering. More specifically, this work proposes a novel framework for performing image clustering from deep embeddings by combining instance-level contrastive learning with a deep embedding based cluster center predictor. Our approach jointly learns representations and predicts cluster centers in an end-to-end manner. This is accomplished via a three-pronged approach that combines a clustering loss, an instance-wise contrastive loss, and an anchor loss. Our fundamental intuition is that using an ensemble loss that incorporates instance-level features and a clustering procedure focusing on semantic similarity reinforces learning better representations in the latent space. We observe that our method performs exceptionally well on popular vision datasets when evaluated using standard clustering metrics such as Normalized Mutual Information (NMI), in addition to producing geometrically well-separated cluster embeddings as defined by the Euclidean distance. Our framework performs on par with widely accepted clustering methods and outperforms the state-of-the-art contrastive learning method on the CIFAR-10 dataset with an NMI score of 0.772, a 7-8% improvement on the strong baseline.

Committee: Ali Jannesari (major professor), Qi Li, Gurpur Prabhu, and John Paul Just

Join on WebExhttps://iastate.webex.com/iastate/j.php?MTID=m9ffadf3c53c539d82613295b9e5078b8