Pith. sign in

Feature Selection for Huge Data via Minipatch Learning

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

Feature selection often leads to increased model interpretability, faster computation, and improved model performance by discarding irrelevant or redundant features. While feature selection is a well-studied problem with many widely-used techniques, there are typically two key challenges: i) many existing approaches become computationally intractable in huge-data settings with millions of observations and features; and ii) the statistical accuracy of selected features degrades in high-noise, high-correlation settings, thus hindering reliable model interpretation. We tackle these problems by proposing Stable Minipatch Selection (STAMPS) and Adaptive STAMPS (AdaSTAMPS). These are meta-algorithms that build ensembles of selection events of base feature selectors trained on many tiny, (adaptively-chosen) random subsets of both the observations and features of the data, which we call minipatches. Our approaches are general and can be employed with a variety of existing feature selection strategies and machine learning techniques. In addition, we provide theoretical insights on STAMPS and empirically demonstrate that our approaches, especially AdaSTAMPS, dominate competing methods in terms of feature selection accuracy and computational time.

citation-role summary

background 1

citation-polarity summary

fields

stat.ML 1

years

2025 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

unclear 1

representative citing papers

iLOCO: Distribution-Free Inference for Feature Interactions

stat.ML · 2025-02-10 · conditional · novelty 5.0

iLOCO estimates the importance of pairwise and higher-order feature interactions in any predictive model and provides distribution-free confidence intervals via data splitting or minipatch ensembles.

citing papers explorer

Showing 1 of 1 citing paper.

  • iLOCO: Distribution-Free Inference for Feature Interactions stat.ML · 2025-02-10 · conditional · none · ref 54 · internal anchor

    iLOCO estimates the importance of pairwise and higher-order feature interactions in any predictive model and provides distribution-free confidence intervals via data splitting or minipatch ensembles.