Identifiable Deep Generative Models via Sparse Decoding

1 code implementation • 20 Oct 2021 • Gemma E. Moran, Dhanya Sridhar, Yixin Wang, David M. Blei

The underlying model is sparse in that each observed feature (i. e. each dimension of the data) depends on a small subset of the latent factors.

Representation Learning

Population Predictive Checks

no code implementations • 2 Aug 2019 • Gemma E. Moran, David M. Blei, Rajesh Ranganath

However, PPCs use the data twice -- both to calculate the posterior predictive and to evaluate it -- which can lead to overconfident assessments of the quality of a model.

Spike-and-Slab Group Lassos for Grouped Regression and Sparse Generalized Additive Models

1 code implementation • 5 Mar 2019 • Ray Bai, Gemma E. Moran, Joseph Antonelli, Yong Chen, Mary R. Boland

We introduce the spike-and-slab group lasso (SSGL) for Bayesian estimation and variable selection in linear regression with grouped variables.

Methodology

Variance prior forms for high-dimensional Bayesian variable selection

1 code implementation • 9 Jan 2018 • Gemma E. Moran, Veronika Rockova, Edward I. George

In a similar way, we show that conjugate priors for linear regression, which induce prior dependence, can lead to such underestimation in the Bayesian high-dimensional regression setting.

Methodology