Cohort Organized Learning: Clustering Through Agreement
For practitioners needing to cluster diverse data types without predefined metrics, CoOL offers a flexible neural network-based approach, though the paper focuses on methodology rather than empirical benchmarks.
CoOL introduces a clustering method that avoids explicit distance or similarity computations by using neural networks trained via expectation maximization. It is demonstrated on vector data and images, showing convergence monitoring and cluster evaluation.
In this article we describe Cohort Organized Learning (CoOL), a method for clustering data without explicit distance or similarity computations. Herein, we will describe CoOL, derive the gradients determined by expectation maximization to train the networks, show how to monitor convergence during training and evaluate the clusters after training, and discuss a series of examples and use cases. We also discuss CoOL's limitations and future prospects on related tasks. Because CoOL uses neural networks to estimate the clusters, it can be used to cluster any data that can be made compatible and we illustrate this on vector data and images.