Disentangling to Cluster: Gaussian Mixture Variational Ladder Autoencoders
arXiv:1909.11501
Abstract
In clustering we normally output one cluster variable for each datapoint. However it is not necessarily the case that there is only one way to partition a given dataset into cluster components. For example, one could cluster objects by their colour, or by their type. Different attributes form a hierarchy, and we could wish to cluster in any of them. By disentangling the learnt latent representations of some dataset into different layers for different attributes we can then cluster in those latent spaces. We call this "disentangled clustering". Extending Variational Ladder Autoencoders (Zhao et al., 2017), we propose a clustering algorithm, VLAC, that outperforms a Gaussian Mixture DGM in cluster accuracy over digit identity on the test set of SVHN. We also demonstrate learning clusters jointly over numerous layers of the hierarchy of latent variables for the data, and show component-wise generation from this hierarchical model.
References in corpus (13)
- Categorical Reparameterization with Gumbel-Softmax
- The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
- A Survey on Multi-view Learning
- Disentangling by Factorising
- Semi-Supervised Learning with Ladder Networks
- Learning Discrete Representations via Information Maximizing Self-Augmented Training
- Towards a Definition of Disentangled Representations
- Joint Unsupervised Learning of Deep Representations and Image Clusters
- Structured Disentangled Representations
- Deconstructing the Ladder Network Architecture
- Nonparametric Variational Auto-encoders for Hierarchical Representation Learning
- Learning Latent Superstructures in Variational Autoencoders for Deep Multidimensional Clustering
- Learning Latent Representations in Neural Networks for Clustering through Pseudo Supervision and Graph-based Activity Regularization