Timezone: »

Missing Data Imputation and Acquisition with Deep Hierarchical Models and Hamiltonian Monte Carlo
Ignacio Peis · Chao Ma · José Miguel Hernández-Lobato

Wed Nov 30 09:00 AM -- 11:00 AM (PST) @ Hall J #114

Variational Autoencoders (VAEs) have recently been highly successful at imputing and acquiring heterogeneous missing data. However, within this specific application domain, existing VAE methods are restricted by using only one layer of latent variables and strictly Gaussian posterior approximations. To address these limitations, we present HH-VAEM, a Hierarchical VAE model for mixed-type incomplete data that uses Hamiltonian Monte Carlo with automatic hyper-parameter tuning for improved approximate inference. Our experiments show that HH-VAEM outperforms existing baselines in the tasks of missing data imputation and supervised learning with missing features. Finally, we also present a sampling-based approach for efficiently computing the information gain when missing features are to be acquired with HH-VAEM. Our experiments show that this sampling-based approach is superior to alternatives based on Gaussian approximations.

Author Information

Ignacio Peis (Universidad Carlos III de Madrid)
Ignacio Peis

I am interested in the connection between Deep Learning and Probabilistic Modelling. My current research lies in creating more expressive generative models, increasing their robustness (e.g. dealing with mixed-type data or missing data) and developing better inference methods. Previously I was a visiting researcher at the Machine Learning Group in the Department of the Engineering, University of Cambridge. I obtained two MSc degrees in Telecommunications Engineering and Signal Processing from UC3M, and my Bachelor degree in Telecommunications Engineering from UGR.

Chao Ma (University of Cambridge)
José Miguel Hernández-Lobato (University of Cambridge)

More from the Same Authors