DeepPINK: reproducible feature selection in deep neural networks

Abstract

Deep learning has become increasingly popular in both supervised and unsupervised machine learning thanks to its outstanding empirical performance. However, because of their intrinsic complexity, most deep learning methods are largely treated as black box tools with little interpretability. Even though recent attempts have been made to facilitate the interpretability of deep neural networks (DNNs), existing methods are susceptible to noise and lack of robustness. Therefore, scientists are justifiably cautious about the reproducibility of the discoveries, which is often related to the interpretability of the underlying statistical models. In this paper, we describe a method to increase the interpretability and reproducibility of DNNs by incorporating the idea of feature selection with controlled error rate. By designing a new DNN architecture and integrating it with the recently proposed knockoffs framework, we perform feature selection with a controlled error rate, while maintaining high power. This new method, DeepPINK (Deep feature selection using Paired-Input Nonlinear Knockoffs), is applied to both simulated and real data sets to demonstrate its empirical utility.

Cited by in corpus (19)

• A User-Guided Bayesian Framework for Ensemble Feature Selection in Life Science Applications (UBayFS)
• Simple stopping criteria for information theoretic feature selection
• A Power Analysis of the Conditional Randomization Test and Knockoffs
• False Discovery Rate Control via Data Splitting
• Discovering Conditionally Salient Features with Statistical Guarantees
• Consistent Feature Selection for Analytic Deep Neural Networks
• Multiple competition-based FDR control for peptide detection
• A Scale-free Approach for False Discovery Rate Control in Generalized Linear Models
• A Bregman Method for Structure Learning on Sparse Directed Acyclic Graphs
• Competition-based control of the false discovery proportion
• More Powerful Selective Kernel Tests for Feature Selection
• Deep Neural Networks Guided Ensemble Learning for Point Estimation
• An Automatic Relevance Determination Prior Bayesian Neural Network for Controlled Variable Selection
• Neural Gaussian Mirror for Controlled Feature Selection in Neural Networks
• DANCE: Enhancing saliency maps using decoys
• Learning to Increase the Power of Conditional Randomization Tests
• $\P$ILCRO: Making Importance Landscapes Flat Again
• Deep-gKnock: nonlinear group-feature selection with deep neural network
• Differentially Private Variable Selection via the Knockoff Filter