Knockoff Statistic

Knockoff statistics provide a powerful framework for feature selection in high-dimensional data, guaranteeing control over the false discovery rate (FDR) – the proportion of falsely identified significant features. Recent research focuses on extending knockoff methods to handle complex data structures, such as time series and nonlinear relationships, often leveraging deep learning architectures like LSTMs and transformers to generate knockoff variables and improve selection power. This approach is proving valuable across diverse fields, enabling more reliable and interpretable analyses in applications ranging from genomics and climate science to reinforcement learning and sales prediction.

Papers

April 5, 2024

DeepLINK-T: deep learning inference for time series data using knockoffs and LSTM
Wenxuan Zuo, Zifan Zhu, Yuxuan Du, Yi-Chun Yeh, Jed A. Fuhrman, Jinchi Lv, Yingying Fan, Fengzhu Sun
Scientific Inference LSTM Network Time Series Data High Dimensional Time Series Longitudinal Dataset Knockoff Statistic Model X Knockoff

February 27, 2024

DeepDRK: Deep Dependency Regularized Knockoff for Feature Selection
Hongyu Shen, Yici Yan, Zhizhen Zhao
Feature Selection Knockoff Statistic Learning Based Generative Model X Knockoff

October 7, 2023

The Conditional Prediction Function: A Novel Technique to Control False Discovery Rate for Complex Models
Yushu Shi, Michael Martens
Complex Model Novel Technique False Discovery Rate Prediction Function Knockoff Statistic Irrelevant Variable

March 24, 2023

Sequential Knockoffs for Variable Selection in Reinforcement Learning
Tao Ma, Hengrui Cai, Zhengling Qi, Chengchun Shi, Eric B. Laber
Reinforcement Learning Markov Decision Process Reinforcement Learning Algorithm Policy OpTimization Variable Selection Knockoff Statistic

September 23, 2022

Time Series Causal Link Estimation under Hidden Confounding using Knockoff Interventions
Violeta Teodora Trifunov, Maha Shadaydeh, Joachim Denzler
Variational Autoencoder Causal Relationship Causal Effect Estimation Latent Confounding Knockoff Statistic Aerosol Cloud

July 8, 2022

Causal Discovery using Model Invariance through Knockoff Interventions
Wasim Ahmad, Maha Shadaydeh, Joachim Denzler
Causal Discovery Granger Causality Causal Prediction Model Invariance Multivariate System Knockoff Statistic Causality Method

March 9, 2022

Error-based Knockoffs Inference for Controlled Feature Selection
Xuebin Zhao, Hong Chen, Yingjie Wang, Weifu Li, Tieliang Gong, Yulong Wang, Feng Zheng
Feature Importance Knockoff Statistic Model X Knockoff

February 26, 2022

Missing Value Knockoffs
Deniz Koyuncu, Bülent Yener
Imputation Algorithm Latent Variable Model Variable Selection Missing Data Imputation Knockoff Statistic