Union of Intersections (UoI) for Interpretable Data Driven Discovery and Prediction

TL;DR

The paper introduces the Union of Intersections (UoI) framework, a scalable method for model selection and estimation that enhances interpretability and prediction accuracy in complex scientific data analysis.

Contribution

The paper presents the UoI framework, a novel modular approach combining intersection and union operations for improved model interpretability and predictive performance.

Findings

UoI_{Lasso} extracts interpretable features from electrophysiology data.

UoI_{L1Logistic} and UoI_{CUR} improve prediction parsimony on biomedical datasets.

UoI-based methods achieve low-variance, nearly unbiased estimation with high prediction accuracy.

Abstract

The increasing size and complexity of scientific data could dramatically enhance discovery and prediction for basic scientific applications. Realizing this potential, however, requires novel statistical analysis methods that are both interpretable and predictive. We introduce Union of Intersections (UoI), a flexible, modular, and scalable framework for enhanced model selection and estimation. Methods based on UoI perform model selection and model estimation through intersection and union operations, respectively. We show that UoI-based methods achieve low-variance and nearly unbiased estimation of a small number of interpretable features, while maintaining high-quality prediction accuracy. We perform extensive numerical investigation to evaluate a UoI algorithm ($UoI_{Lasso}$) on synthetic and real data. In doing so, we demonstrate the extraction of interpretable functional networks from human electrophysiology recordings as well as accurate prediction of phenotypes from genotype-phenotype data with reduced features. We also show (with the $UoI_{L1Logistic}$ and $UoI_{CUR}$ variants of the basic framework) improved prediction parsimony for classification and matrix factorization on several benchmark biomedical data sets. These results suggest that methods based on the UoI framework could improve interpretation and prediction in data-driven discovery across scientific fields.