RandALO: Out-of-sample risk estimation in no time flat

Autor:	Nobel, Parth T., LeJeune, Daniel, Candès, Emmanuel J.
Rok vydání:	2024
Předmět:	Mathematics - Statistics Theory Computer Science - Machine Learning Mathematics - Optimization and Control Statistics - Computation Statistics - Machine Learning
Druh dokumentu:	Working Paper
Popis:	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. Comment: 25 pages, 9 figures
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2409.09781 Zobrazit plný text záznamu View this record from Arxiv