back to list

Project: Topics in Deep Clustering


Deep clustering is a well-researched field with promising approaches. Traditional nonconvex clustering methods require the definition of a kernel matrix, whose parameters vastly influence the result, and are hence difficult to specify. In turn, the promise of deep clustering is that a feature transformation can be learned automatically, based on the data, such that points in the transformed feature space are easy to cluster.

However, as often with deep neural networks, making the network learn sensible representations is difficult. In this case, we would like to find representations that clearly separate the clusters. If you are interested in looking into the state-of-the art approaches and finding out how we can learn the representations that we want, we can discuss possible approaches to tackle this task. If you want to take on this topic, you have ideally followed the course 2IIG0 or JBI030, such that you know the fundamentals of k-means and matrix factorization. 

Here is some literature to get inspired:

[1] Schnellbach, Janik and Márton Kajó. “Clustering with Deep Neural Networks – An Overview of Recent Methods.” (2020).

[2] Nutakki, Gopi Chand et al. “An Introduction to Deep Clustering.” Clustering Methods for Big Data Analytics (2018):

Sibylle Hess
Get in contact