Model Monitoring in the Absence of Labeled Data via Feature Attributions Distributions

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autore principale: Mougan, Carlos
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913666474967040
author Mougan, Carlos
author_facet Mougan, Carlos
contents Model monitoring involves analyzing AI algorithms once they have been deployed and detecting changes in their behaviour. This thesis explores machine learning model monitoring ML before the predictions impact real-world decisions or users. This step is characterized by one particular condition: the absence of labelled data at test time, which makes it challenging, even often impossible, to calculate performance metrics. The thesis is structured around two main themes: (i) AI alignment, measuring if AI models behave in a manner consistent with human values and (ii) performance monitoring, measuring if the models achieve specific accuracy goals or desires. The thesis uses a common methodology that unifies all its sections. It explores feature attribution distributions for both monitoring dimensions. Using these feature attribution explanations, we can exploit their theoretical properties to derive and establish certain guarantees and insights into model monitoring.
format Preprint
id arxiv_https___arxiv_org_abs_2501_10774
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Model Monitoring in the Absence of Labeled Data via Feature Attributions Distributions
Mougan, Carlos
Machine Learning
Model monitoring involves analyzing AI algorithms once they have been deployed and detecting changes in their behaviour. This thesis explores machine learning model monitoring ML before the predictions impact real-world decisions or users. This step is characterized by one particular condition: the absence of labelled data at test time, which makes it challenging, even often impossible, to calculate performance metrics. The thesis is structured around two main themes: (i) AI alignment, measuring if AI models behave in a manner consistent with human values and (ii) performance monitoring, measuring if the models achieve specific accuracy goals or desires. The thesis uses a common methodology that unifies all its sections. It explores feature attribution distributions for both monitoring dimensions. Using these feature attribution explanations, we can exploit their theoretical properties to derive and establish certain guarantees and insights into model monitoring.
title Model Monitoring in the Absence of Labeled Data via Feature Attributions Distributions
topic Machine Learning
url https://arxiv.org/abs/2501.10774