Contributed Papers

Advances in Sparse Learning

Wed, Aug 6, 8:30 AM - 10:20 AM Room CC-105B Music City Center

Section on Statistical Learning and Data Science

7 Presentations

More of Less: A Rashomon Algorithm for Sparse Model Sets

8:35 AM - 8:50 AM

Co-authors: Roberto Molinari (Auburn University), Roberto Molinari (Auburn University), Cesare Miglioli (Purdue University), Gaetan Bakalli (Emlyon Business School), Stephane Guerrier, Samuel Orso

SWAG: Sparse Wrapper Algorithm Latent Representations Replication Crisis High-Dimensional Problems Variable Importance Explainable AI

An augmented pseudo likelihood approach for multistate modeling in cross-sectional studies

8:50 AM - 9:05 AM

Co-authors: Fangya Mao, Fangya Mao, Nicole Campos (Center for Health Decision Science, Harvard T.H. Chan School of Public Health), Li Cheung (National Cancer Institute)

multistate current status data portable natural history models recurrent processes data integration semi-Markov models

Joint Probability Estimation of Many Binary Outcomes via Localized Adversarial Lasso

9:05 AM - 9:20 AM

Co-authors: Alexandre Belloni (Duke University), Yan Chen (Duke University), Matthew Harding (University of California Irvine)

many binary classification penalized regression adversarial lasso

Lasso rates hold even after randomly discarding most of the predictors

9:20 AM - 9:35 AM

Co-authors: Salvador Balkus (Harvard T.H. Chan School of Public Health), Salvador Balkus (Harvard T.H. Chan School of Public Health), Nima Hejazi (Harvard T.H. Chan School of Public Health)

lasso function estimation empirical risk minimization statistical learning high-dimensional statistics

A Statistical Test for SWAG: Assessing the Rashomon Effect in Sparse Model Selection

9:35 AM - 9:50 AM

Co-authors: Yagmur Yavuzozdemir, Roberto Molinari (Auburn University), Yagmur Yavuzozdemir

SWAG Multi-model selection Statistical testing Rashomon effect Sparse models

The Predictive Degrees of Freedom of LASSO

9:50 AM - 10:05 AM

Co-authors: Xuerong Wang (The Ohio State University), Yoonkyung Lee (The Ohio State University), Xuerong Wang (The Ohio State University)

Double Descent Infinitesimal Jackknife Lasso Model Degrees of Freedom Variance Estimation

Tree-guided equi-sparisty pursuit for high-dimensional regression and classification

10:05 AM - 10:20 AM

Co-authors: Jinwen Fu, Aaron Molstad (University of Minnesota), Jinwen Fu, Hui Zou (University of Minnesota)

feature aggregation equi-sparsity tree-guided regularization high-dimensional linear models post-selection inference proximal operator