English

VAEs in the Presence of Missing Data

Machine Learning 2021-03-23 v3 Machine Learning

Abstract

Real world datasets often contain entries with missing elements e.g. in a medical dataset, a patient is unlikely to have taken all possible diagnostic tests. Variational Autoencoders (VAEs) are popular generative models often used for unsupervised learning. Despite their widespread use it is unclear how best to apply VAEs to datasets with missing data. We develop a novel latent variable model of a corruption process which generates missing data, and derive a corresponding tractable evidence lower bound (ELBO). Our model is straightforward to implement, can handle both missing completely at random (MCAR) and missing not at random (MNAR) data, scales to high dimensional inputs and gives both the VAE encoder and decoder principled access to indicator variables for whether a data element is missing or not. On the MNIST and SVHN datasets we demonstrate improved marginal log-likelihood of observed data and better missing data imputation, compared to existing approaches.

Keywords

Cite

@article{arxiv.2006.05301,
  title  = {VAEs in the Presence of Missing Data},
  author = {Mark Collier and Alfredo Nazabal and Christopher K. I. Williams},
  journal= {arXiv preprint arXiv:2006.05301},
  year   = {2021}
}

Comments

Accepted to ICML Workshop on the Art of Learning with Missing Values (Artemiss), 17 July 2020

R2 v1 2026-06-23T16:10:53.246Z