Deep Unsupervised Clustering with Gaussian Mixture Variational Autoencoders
This work addresses unsupervised clustering for data analysis, but it is incremental as it adapts an existing heuristic to a variant model.
The paper tackles unsupervised clustering using a Gaussian mixture variational autoencoder, addressing over-regularization and cluster degeneracy with a minimum information constraint heuristic. It demonstrates competitive performance on synthetic data, MNIST, and SVHN, achieving distinct and interpretable clusters.
We study a variant of the variational autoencoder model (VAE) with a Gaussian mixture as a prior distribution, with the goal of performing unsupervised clustering through deep generative models. We observe that the known problem of over-regularisation that has been shown to arise in regular VAEs also manifests itself in our model and leads to cluster degeneracy. We show that a heuristic called minimum information constraint that has been shown to mitigate this effect in VAEs can also be applied to improve unsupervised clustering performance with our model. Furthermore we analyse the effect of this heuristic and provide an intuition of the various processes with the help of visualizations. Finally, we demonstrate the performance of our model on synthetic data, MNIST and SVHN, showing that the obtained clusters are distinct, interpretable and result in achieving competitive performance on unsupervised clustering to the state-of-the-art results.