Leveraging variational autoencoders for multiple data imputation

Breeshey Roskams-Hieter, Jude Daniel Wells, Sara Wade

Keywords: Variational autoencoders, missing data, multiple imputation, deep learning, power likelihood

TL;DR: Multiple data imputation using variational autoencoders allows accurate imputation of missing data, while retaining good coverage.

Abstract: Missing data persists as a major barrier to data analysis across numerous applications. Recently, deep generative models have been used for imputation of missing data, motivated by their ability to capture highly non-linear and complex relationships in the data. In this work, we investigate the ability of deep models, namely variational autoencoders (VAEs), to account for uncertainty in missing data through multiple imputation strategies. We find that VAEs provide poor empirical coverage of missing data, with underestimation and overconfident imputations, particularly for more extreme missing data values. To overcome this, we employ $\beta$-VAEs, which viewed from a generalized Bayes framework, provide robustness to model misspecification. Assigning a good value of $\beta$ is critical for uncertainty calibration and we demonstrate how this can be achieved using cross-validation. In downstream tasks, we show how multiple imputation with $\beta$-VAEs can avoid false discoveries that arise as artefacts of imputation.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Generative models

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/leveraging-variational-autoencoders-for/code)