We consider the problem of handling missing data with deep latent variable models
(DLVMs). First, we present a simple technique to train DLVMs when the training set contains …

Consider the problem of imputing missing values in a dataset. One the one hand,
conventional approaches using iterative imputation benefit from the simplicity and …

In spite of remarkable progress in deep latent variable generative modeling, training still
remains a challenge due to a combination of optimization and generalization issues. In …

Ensuring data quality in biomedical sciences is crucial for reliable research outcomes,
particularly as precision medicine continues to gain prominence. Missing values …

Variational autoencoders (VAEs) provide an effective and simple method for modeling
complex distributions. However, training VAEs often requires considerable hyperparameter …

In narrow asymptotic settings Gaussian VAE models of continuous data have been shown to
possess global optima aligned with ground-truth distributions. Even so, it is well known that …

When a missing process depends on the missing values themselves, it needs to be explicitly
modelled and taken into account while doing likelihood-based inference. We present an …

Likelihood-based, or explicit, deep generative models use neural networks to construct
flexible high-dimensional densities. This formulation directly contradicts the manifold …

The issue of missing data in supervised learning has been largely overlooked, especially in
the deep learning community. We investigate strategies to adapt neural architectures for …

We propose and investigate new complementary methodologies for estimating predictive
variance networks in regression neural networks. We derive a locally aware mini-batching …