GP-VAE: Deep Probabilistic Time Series Imputation

By Vincent Fortuin, Dmitry Baranchuk, Gunnar Rätsch, Stephan Mandt. International Conference on Artificial Intelligence and Statistics, 2019

402 citations recorded by the source corpus.

Multivariate time series with missing values are common in areas such as healthcare and finance, and have grown in number and complexity over the years. This raises the question whether deep learning methodologies can outperform classical data imputation methods in this domain. However, naive applications of deep learning fall short in giving reliable confidence estimates and lack interpretability. We propose a new deep sequential latent variable model for dimensionality reduction and data imputation. Our modeling assumption is simple and interpretable: the high dimensional time series has a lower-dimensional representation which evolves smoothly in time according to a Gaussian process. The non-linear dimensionality reduction in the presence of missing data is achieved using a VAE approach with a novel structured variational approximation. We demonstrate that our approach outperforms several classical and deep learning-based data imputation methods on high-dimensional data from the domains of computer vision and healthcare, while additionally improving the smoothness of the imputations and providing interpretable uncertainty estimates.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/820b8c5e2630e663c191467a7654a377a83cdc55

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

Citations in the corpus

Diffusion-based Time Series Imputation and Forecasting with Structured State Space Models (2022, methodology): “This includes GAN-based approaches such as E2-GAN [34], and GRUI-GAN [33] or VAE-based approaches such as GP-VAE [15].”
Dynamical Variational Autoencoders: A Comprehensive Review (2020, background): “GP-VAE: In the recent paper (Fortuin et al., 2020), the authors combine a VAE for observed data dimension reduction and a multivariate Gaussian Process (GP) (Williams and Rasmussen, 2006) for modeling the dynamics of the In the supplementary material to the paper, they break the dependency between z”
Continual Multi-task Gaussian Processes (2019, methodology): “Moreover, the continual single-output GP model could be used as a latent baseline in the multivariate time series imputation method of Fortuin et al. (2019), which uses a GP to capture temporal dependencies between real-valued latent variables that are later connected to a deep sequential…”
Diffusion Models: A Comprehensive Survey of Methods and Applications (2022, background): “Recent years, imputation methods have been greatly for both deterministic imputation [23, 27, 154] and probabilistic imputation [65], including diffusion-based approaches.”