An Automatic Relevance Determination Prior Bayesian Neural Network for Controlled Variable Selection

01/06/2020

∙

We present an Automatic Relevance Determination prior Bayesian Neural Network(BNN-ARD) weight l2-norm measure as a feature importance statistic for the model-x knockoff filter. We show on both simulated data and the Norwegian wind farm dataset that the proposed feature importance statistic yields statistically significant improvements relative to similar feature importance measures in both variable selection power and predictive performance on a real world dataset.

READ FULL TEXT

An Automatic Relevance Determination Prior Bayesian Neural Network for Controlled Variable Selection

Model selection for Gaussian processes utilizing sensitivity of posterior predictive distribution

Differentially Private Variable Selection via the Knockoff Filter

Automatic Relevance Determination Bayesian Neural Networks for Credit Card Default Modelling

Laplacian Prior Variational Automatic Relevance Determination for Transmission Tomography

Variable Selection with Rigorous Uncertainty Quantification using Deep Bayesian Neural Networks: Posterior Concentration and Bernstein-von Mises Phenomenon

Comparative Study of Inference Methods for Interpolative Decomposition

Inconsistency identification in network meta-analysis via stochastic search variable selection

An Automatic Relevance Determination Prior Bayesian Neural Network for Controlled Variable Selection

Related Research

Model selection for Gaussian processes utilizing sensitivity of posterior predictive distribution

Differentially Private Variable Selection via the Knockoff Filter

Automatic Relevance Determination Bayesian Neural Networks for Credit Card Default Modelling

Laplacian Prior Variational Automatic Relevance Determination for Transmission Tomography

Variable Selection with Rigorous Uncertainty Quantification using Deep Bayesian Neural Networks: Posterior Concentration and Bernstein-von Mises Phenomenon

Comparative Study of Inference Methods for Interpolative Decomposition

Inconsistency identification in network meta-analysis via stochastic search variable selection