Handling Incomplete Heterogeneous Data using VAEs

By Alfredo Nazábal, Pablo M. Olmos, Zoubin Ghahramani, Isabel Valera. Pattern Recognition, 2018

449 citations recorded by the source corpus.

Variational autoencoders (VAEs), as well as other generative models, have been shown to be efficient and accurate for capturing the latent structure of vast amounts of complex high-dimensional data. However, existing VAEs can still not directly handle data that are heterogenous (mixed continuous and discrete) or incomplete (with missing data at random), which is indeed common in real-world applications. In this paper, we propose a general framework to design VAEs suitable for fitting incomplete heterogenous data. The proposed HI-VAE includes likelihood models for real-valued, positive real valued, interval, categorical, ordinal and count data, and allows accurate estimation (and potentially imputation) of missing data. Furthermore, HI-VAE presents competitive predictive performance in supervised tasks, outperforming supervised models when trained on incomplete data.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing

View Paper: https://www.semanticscholar.org/paper/a9597f3bb790c607ce0a87277523226feac5a8d7

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

TabDDPM: Modelling Tabular Data with Diffusion Models (2022, background): “As for numerical features, the possible extension of TabDDPM can be inspired by (Nazabal et al., 2020) that distinguish different types of numerical variables, i.e., real-valued, positive real-valued or ordinal.”
MIWAE: Deep Generative Modelling and Imputation of Incomplete Data Sets (2019, methodology): “One of the DLVM was trained using the MVAE bound, which is equivalent to the objective of Nazabal et al. (2018), and one using the MIWAE bound with K = 20.”
Towards Realistic Individual Recourse and Actionable Explanations in Black-Box Decision Making Systems (2019, methodology): “Datatype settings for defaultCredit dataset to train the HI-VAE: HI-VAE settings Settings for training the HI-VAE from (Nazabal et al., 2018) are provided in Table 8 Decision Making Systems Classiﬁers tested for recourse on defaultCredit dataset.”
GP-VAE: Deep Probabilistic Time Series Imputation (2019, methodology): “We compare our model against a standard VAE (no latent GP and standard ELBO over all features), the HI-VAE [37], as well as mean imputation and forward imputation.”
Deep Neural Networks and Tabular Data: A Survey (2021, methodology): “To this end, these methods primarily use (tabular data) variational autoencoders [142], [185].”
Surgical data science – from concepts toward clinical translation (2020, background): “Other work has specifically focused on handling incomplete heterogeneous data with Variational Autoencoders (VAEs) (Nazábal et al., 2020).”
A survey of algorithmic recourse: definitions, formulations, solutions, and prospects (2020, methodology): “ltiple categories are be simultaneously active, and thermometer-encoded ordinal variables are required to satisfy � �,�≥� +1 ∀� ∈{0,1}. For a detailed overview, we refer to the work of Nazabal et al. [135]. In addition to ! tabular data, one may require contrastive explanations for image-based or pt”