Fast Leave-One-Out Approximation from Fragment-Target Prevalence Vectors (molFTP) : From Dummy Masking to Key-LOO for Leakage-Free Feature Construction

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autor principal: Godin, Guillaume
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918155916410880
author Godin, Guillaume
author_facet Godin, Guillaume
contents We introduce molFTP (molecular fragment-target prevalence), a compact representation that delivers strong predictive performance. To prevent feature leakage across cross-validation folds, we implement a dummy-masking procedure that removes information about fragments present in the held-out molecules. We further show that key leave-one-out (key-loo) closely approximates true molecule-level leave-one-out (LOO), with deviation below 8% on our datasets. This enables near full data training while preserving unbiased cross-validation estimates of model performance. Overall, molFTP provides a fast, leakage-resistant fragment-target prevalence vectorization with practical safeguards (dummy masking or key-LOO) that approximate LOO at a fraction of its cost.
format Preprint
id arxiv_https___arxiv_org_abs_2510_06029
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Fast Leave-One-Out Approximation from Fragment-Target Prevalence Vectors (molFTP) : From Dummy Masking to Key-LOO for Leakage-Free Feature Construction
Godin, Guillaume
Machine Learning
Artificial Intelligence
We introduce molFTP (molecular fragment-target prevalence), a compact representation that delivers strong predictive performance. To prevent feature leakage across cross-validation folds, we implement a dummy-masking procedure that removes information about fragments present in the held-out molecules. We further show that key leave-one-out (key-loo) closely approximates true molecule-level leave-one-out (LOO), with deviation below 8% on our datasets. This enables near full data training while preserving unbiased cross-validation estimates of model performance. Overall, molFTP provides a fast, leakage-resistant fragment-target prevalence vectorization with practical safeguards (dummy masking or key-LOO) that approximate LOO at a fraction of its cost.
title Fast Leave-One-Out Approximation from Fragment-Target Prevalence Vectors (molFTP) : From Dummy Masking to Key-LOO for Leakage-Free Feature Construction
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2510.06029