Saved in:
Bibliographic Details
Main Authors: Shen, Hongyu, Yan, Yici, Zhao, Zhizhen
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.17176
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929582655930368
author Shen, Hongyu
Yan, Yici
Zhao, Zhizhen
author_facet Shen, Hongyu
Yan, Yici
Zhao, Zhizhen
contents Model-X knockoff has garnered significant attention among various feature selection methods due to its guarantees for controlling the false discovery rate (FDR). Since its introduction in parametric design, knockoff techniques have evolved to handle arbitrary data distributions using deep learning-based generative models. However, we have observed limitations in the current implementations of the deep Model-X knockoff framework. Notably, the "swap property" that knockoffs require often faces challenges at the sample level, resulting in diminished selection power. To address these issues, we develop "Deep Dependency Regularized Knockoff (DeepDRK)," a distribution-free deep learning method that effectively balances FDR and power. In DeepDRK, we introduce a novel formulation of the knockoff model as a learning problem under multi-source adversarial attacks. By employing an innovative perturbation technique, we achieve lower FDR and higher power. Our model outperforms existing benchmarks across synthetic, semi-synthetic, and real-world datasets, particularly when sample sizes are small and data distributions are non-Gaussian.
format Preprint
id arxiv_https___arxiv_org_abs_2402_17176
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DeepDRK: Deep Dependency Regularized Knockoff for Feature Selection
Shen, Hongyu
Yan, Yici
Zhao, Zhizhen
Machine Learning
68T07
I.5.1
Model-X knockoff has garnered significant attention among various feature selection methods due to its guarantees for controlling the false discovery rate (FDR). Since its introduction in parametric design, knockoff techniques have evolved to handle arbitrary data distributions using deep learning-based generative models. However, we have observed limitations in the current implementations of the deep Model-X knockoff framework. Notably, the "swap property" that knockoffs require often faces challenges at the sample level, resulting in diminished selection power. To address these issues, we develop "Deep Dependency Regularized Knockoff (DeepDRK)," a distribution-free deep learning method that effectively balances FDR and power. In DeepDRK, we introduce a novel formulation of the knockoff model as a learning problem under multi-source adversarial attacks. By employing an innovative perturbation technique, we achieve lower FDR and higher power. Our model outperforms existing benchmarks across synthetic, semi-synthetic, and real-world datasets, particularly when sample sizes are small and data distributions are non-Gaussian.
title DeepDRK: Deep Dependency Regularized Knockoff for Feature Selection
topic Machine Learning
68T07
I.5.1
url https://arxiv.org/abs/2402.17176