Guardado en:
Detalles Bibliográficos
Autores principales: Lämmle, Sven, Bogoclu, Can, Voßhall, Robert, Haselhoff, Anselm, Roos, Dirk
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2406.07474
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910490200899584
author Lämmle, Sven
Bogoclu, Can
Voßhall, Robert
Haselhoff, Anselm
Roos, Dirk
author_facet Lämmle, Sven
Bogoclu, Can
Voßhall, Robert
Haselhoff, Anselm
Roos, Dirk
contents Real-world applications of machine learning models are often subject to legal or policy-based regulations. Some of these regulations require ensuring the validity of the model, i.e., the approximation error being smaller than a threshold. A global metric is generally too insensitive to determine the validity of a specific prediction, whereas evaluating local validity is costly since it requires gathering additional data.We propose learning the model error to acquire a local validity estimate while reducing the amount of required data through active learning. Using model validation benchmarks, we provide empirical evidence that the proposed method can lead to an error model with sufficient discriminative properties using a relatively small amount of data. Furthermore, an increased sensitivity to local changes of the validity bounds compared to alternative approaches is demonstrated.
format Preprint
id arxiv_https___arxiv_org_abs_2406_07474
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Quantifying Local Model Validity using Active Learning
Lämmle, Sven
Bogoclu, Can
Voßhall, Robert
Haselhoff, Anselm
Roos, Dirk
Machine Learning
Real-world applications of machine learning models are often subject to legal or policy-based regulations. Some of these regulations require ensuring the validity of the model, i.e., the approximation error being smaller than a threshold. A global metric is generally too insensitive to determine the validity of a specific prediction, whereas evaluating local validity is costly since it requires gathering additional data.We propose learning the model error to acquire a local validity estimate while reducing the amount of required data through active learning. Using model validation benchmarks, we provide empirical evidence that the proposed method can lead to an error model with sufficient discriminative properties using a relatively small amount of data. Furthermore, an increased sensitivity to local changes of the validity bounds compared to alternative approaches is demonstrated.
title Quantifying Local Model Validity using Active Learning
topic Machine Learning
url https://arxiv.org/abs/2406.07474