Variational Selective Autoencoder: Learning from Partially-Observed Heterogeneous Data
Yu Gong, Hossein Hajimirsadeghi, Jiawei He, Thibaut Durand, Greg Mori
Abstract
Learning from heterogeneous data poses challenges such as combining data from various sources and of different types. Meanwhile, heterogeneous data are often associated with missingness in real-world applications due to heterogeneity and noise of input sources. In this work, we propose the variational selective autoencoder (VSAE), a general framework to learn representations from partially-observed heterogeneous data. VSAE learns the latent dependencies in heterogeneous data by modeling the joint distribution of observed data, unobserved data, and the imputation mask which represents how the data are missing. It results in a unified model for various downstream tasks including data generation and imputation. Evaluation on both low-dimensional and high-dimensional heterogeneous datasets for these two tasks shows improvement over state-of-the-art models.
BibTeX
@InProceedings{pmlr-v130-gong21a,
title = { Variational Selective Autoencoder: Learning from Partially-Observed Heterogeneous Data },
author = {Gong, Yu and Hajimirsadeghi, Hossein and He, Jiawei and Durand, Thibaut and Mori, Greg},
booktitle = {Proceedings of The 24th International Conference on Artificial Intelligence and Statistics},
pages = {2377--2385},
year = {2021},
editor = {Banerjee, Arindam and Fukumizu, Kenji},
volume = {130},
series = {Proceedings of Machine Learning Research},
month = {13--15 Apr},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v130/gong21a/gong21a.pdf},
url = {https://proceedings.mlr.press/v130/gong21a.html},
abstract = { Learning from heterogeneous data poses challenges such as combining data from various sources and of different types. Meanwhile, heterogeneous data are often associated with missingness in real-world applications due to heterogeneity and noise of input sources. In this work, we propose the variational selective autoencoder (VSAE), a general framework to learn representations from partially-observed heterogeneous data. VSAE learns the latent dependencies in heterogeneous data by modeling the joint distribution of observed data, unobserved data, and the imputation mask which represents how the data are missing. It results in a unified model for various downstream tasks including data generation and imputation. Evaluation on both low-dimensional and high-dimensional heterogeneous datasets for these two tasks shows improvement over state-of-the-art models. }
}