End-to-end Differentiable Clustering with Associative Memories

Bishwajit Saha, Dmitry Krotov, Mohammed J Zaki, Parikshit Ram

Keywords: Clustering, associative memory, self-supervised learning, end-to-end differentiable.

TL;DR: We present a flexible, end-to-end differentiable, associative memory based clustering algorithm which demonstrates strong performance against existing standard baseline clustering schemes.

Abstract: Clustering is a widely used unsupervised learning technique involving an intensive discrete optimization problem. Associative Memory models or AMs are differentiable neural networks defining a recursive dynamical system, which have been integrated with various deep learning architectures. We uncover a novel connection between the AM dynamics and the inherent discrete assignment necessary in clustering to propose a novel unconstrained continuous relaxation of the discrete clustering problem, enabling end-to-end differentiable clustering with AM, dubbed ClAM. Leveraging the pattern completion ability of AMs, we further develop a novel self-supervised clustering loss. Our evaluations on varied datasets demonstrate that ClAM benefits from the self-supervision, and significantly improves upon both the traditional Lloyd's k-means algorithm, and more recent continuous clustering relaxations (by upto 60\% in terms of the Silhouette Coefficient).

Submission Number: 68