PreprintarXiv licencearXiv
On the Duality between Feature and Sample Screening
Abstract
Feature and sample screening reduce the cost of machine learning by eliminating irrelevant features and noninformative samples, respectively. Although recognized as primal-dual counterparts, their relationship remains informal and model-dependent. Viewing screening and duality as transformations of objective functions, we introduce Fenchel-Rockafellar (FR) representations, a class of convex problems encompassing the Lasso and SVM that is closed under both transformations. We then prove that feature and sample screening form an equivariant pair: dualization followed by feature screening is equal to sample screening followed by dualization.
Related
- Finite-sample nonparametric mean tests: Leave-one-out duality and asymptotic optimality
- Exact Dynamics and Finite-Sample Trajectory Recovery of Linear Recursive Feature Machines
- ARISE: An adaptive residual-informed stability ensemble for feature selection in small-sample biomedical omics
- Two-sample inference for sequence-derived feature distributions from deep sequencing data
- Multicollinearity-agnostic feature screening for non-Euclidean responses: a factor adjusted approach
BibTeX
@misc{tran2026duality,
title = {{On the Duality between Feature and Sample Screening}},
author = {Thu-Le Tran},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.02418},
eprint = {2609.02418},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.02418},
}
RIS
TY - UNPB
TI - On the Duality between Feature and Sample Screening
AU - Thu-Le Tran
PY - 2026
DA - 2026/09/02
PB - arXiv
DO - 10.48550/arxiv.2609.02418
UR - https://arxiv.org/abs/2609.02418
AB - Feature and sample screening reduce the cost of machine learning by eliminating irrelevant features and noninformative samples, respectively. Although recognized as primal-dual counterparts, their relationship remains informal and model-dependent. Viewing screening and duality as transformations of objective functions, we introduce Fenchel-Rockafellar (FR) representations, a class of convex problems encompassing the Lasso and SVM that is closed under both transformations. We then prove that feature and sample screening form an equivariant pair: dualization followed by feature screening is equal to sample screening followed by dualization.
ER -