False discovery rate control studies how to select statistically supported findings while limiting the expected proportion of selections that are false. Classical procedures such as Benjamini–Hochberg are joined by knockoff filters, conformal p-values, e-values, and data-splitting methods, which can account for structured or dependent hypotheses. Recent work applies these tools to feature selection, anomaly and out-of-distribution detection, scientific discovery, and sequential or adaptive decisions. A central challenge is preserving reliable guarantees under dependence, model misspecification, or repeated selection without making procedures so conservative that they miss useful discoveries.
McCombs School of Business, The University of Texas at Austin, Austin, TX 78712 · Naveen Jindal School of Management, The University of Texas at Dallas, Richardson, TX 75080 · Warrington College of Business, University of Florida, Gainesville, FL 32611
aShenzhen Key Laboratory of Safety and Security for Next Generation of Industrial Internet, Southern University of Science and Technology, Shenzhen, China · bDepartment of Statistics and Data Science, Southern University of Science and Technology, Shenzhen, China · cSchool of Engineering and Applied Science, University of Pennsylvania, Philadelphia, USA
Department of Statistics and Data Science Southern University of Science and Technology Shenzhen 518055, China · School of Artificial Intelligence The Chinese University of Hong Kong, Shenzhen Shenzhen Loop Area Institute Shenzhen 518172, China
Computer Science and Engineering Sardar Patel Institute of Technology Mumbai, India · Computer Engineering Sardar Patel Institute of Technology Mumbai, India