Learning latent superstructures in variational autoencoders for deep multidimensional clustering

Xiaopeng LI, Zhourong CHEN, Kin Man POON, Nevin L. ZHANG

Research output: Contribution to conferencePoster

15 Citations (Scopus)

Abstract

We investigate a variant of variational autoencoders where there is a superstructure of discrete latent variables on top of the latent features. In general, our superstructure is a tree structure of multiple super latent variables and it is automatically learned from data. When there is only one latent variable in the superstructure, our model reduces to one that assumes the latent features to be generated from a Gaussian mixture model. We call our model the latent tree variational autoencoder (LTVAE). Whereas previous deep learning methods for clustering produce only one partition of data, LTVAE produces multiple partitions of data, each being given by one super latent variable. This is desirable because high dimensional data usually have many different natural facets and can be meaningfully partitioned in multiple ways. Copyright © 2019 ICLR.
Original languageEnglish
Publication statusPublished - May 2019
EventThe Seventh International Conference on Learning Representations - New Orleans, United States
Duration: 06 May 201909 May 2019
https://iclr.cc/Conferences/2019

Conference

ConferenceThe Seventh International Conference on Learning Representations
Abbreviated titleICLR 2019
Country/TerritoryUnited States
CityNew Orleans
Period06/05/1909/05/19
Internet address

Citation

Li, X., Chen, Z., Poon, L. K. M., & Zhang, N. L. (2019, May). Learning latent superstructures in variational autoencoders for deep multidimensional clustering. Poster presented at the Seventh International Conference on Learning Representations (ICLR 2019), Ernest N. Morial Convention Center, New Orleans, US.

Keywords

  • Latent tree model
  • Variational autoencoder
  • Deep learning
  • Latent variable model
  • Bayesian network
  • Structure learning
  • Stepwise EM
  • Message passing
  • Graphical model
  • Multidimensional clustering
  • Unsupervised learning

Fingerprint

Dive into the research topics of 'Learning latent superstructures in variational autoencoders for deep multidimensional clustering'. Together they form a unique fingerprint.