Saved in:
Bibliographic Details
Main Authors: Nobel, Parth, LeJeune, Daniel, Candès, Emmanuel J.
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2409.09781
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912345622577152
author Nobel, Parth
LeJeune, Daniel
Candès, Emmanuel J.
author_facet Nobel, Parth
LeJeune, Daniel
Candès, Emmanuel J.
contents Estimating out-of-sample risk for models trained on large high-dimensional datasets is an expensive but essential part of the machine learning process, enabling practitioners to optimally tune hyperparameters. Cross-validation (CV) serves as the de facto standard for risk estimation but poorly trades off high bias ($K$-fold CV) for computational cost (leave-one-out CV). We propose a randomized approximate leave-one-out (RandALO) risk estimator that is not only a consistent estimator of risk in high dimensions but also less computationally expensive than $K$-fold CV. We support our claims with extensive simulations on synthetic and real data and provide a user-friendly Python package implementing RandALO available on PyPI as randalo and at https://github.com/cvxgrp/randalo.
format Preprint
id arxiv_https___arxiv_org_abs_2409_09781
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle RandALO: Out-of-sample risk estimation in no time flat
Nobel, Parth
LeJeune, Daniel
Candès, Emmanuel J.
Statistics Theory
Machine Learning
Optimization and Control
Computation
Estimating out-of-sample risk for models trained on large high-dimensional datasets is an expensive but essential part of the machine learning process, enabling practitioners to optimally tune hyperparameters. Cross-validation (CV) serves as the de facto standard for risk estimation but poorly trades off high bias ($K$-fold CV) for computational cost (leave-one-out CV). We propose a randomized approximate leave-one-out (RandALO) risk estimator that is not only a consistent estimator of risk in high dimensions but also less computationally expensive than $K$-fold CV. We support our claims with extensive simulations on synthetic and real data and provide a user-friendly Python package implementing RandALO available on PyPI as randalo and at https://github.com/cvxgrp/randalo.
title RandALO: Out-of-sample risk estimation in no time flat
topic Statistics Theory
Machine Learning
Optimization and Control
Computation
url https://arxiv.org/abs/2409.09781