Robust Analysis for Resilient AI System

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Yu, Jin, Ran, Kang, Lulu
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918494302371840
author Wang, Yu
Jin, Ran
Kang, Lulu
author_facet Wang, Yu
Jin, Ran
Kang, Lulu
contents Operational hazards in Manufacturing Industrial Internet (MII) systems generate severe data outliers that cripple traditional statistical analysis. This paper proposes a novel robust regression method, DPD-Lasso, which integrates Density Power Divergence with Lasso regularization to analyze contaminated data from AI resilience experiments. We develop an efficient iterative algorithm to overcome previous computational bottlenecks. Applied to an MII testbed for Aerosol Jet Printing, DPD-Lasso provides reliable, stable performance on both clean and outlier-contaminated data, accurately quantifying hazard impacts. This work establishes robust regression as an essential tool for developing and validating resilient industrial AI systems.
format Preprint
id arxiv_https___arxiv_org_abs_2509_06172
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Robust Analysis for Resilient AI System
Wang, Yu
Jin, Ran
Kang, Lulu
Applications
Machine Learning
Operational hazards in Manufacturing Industrial Internet (MII) systems generate severe data outliers that cripple traditional statistical analysis. This paper proposes a novel robust regression method, DPD-Lasso, which integrates Density Power Divergence with Lasso regularization to analyze contaminated data from AI resilience experiments. We develop an efficient iterative algorithm to overcome previous computational bottlenecks. Applied to an MII testbed for Aerosol Jet Printing, DPD-Lasso provides reliable, stable performance on both clean and outlier-contaminated data, accurately quantifying hazard impacts. This work establishes robust regression as an essential tool for developing and validating resilient industrial AI systems.
title Robust Analysis for Resilient AI System
topic Applications
Machine Learning
url https://arxiv.org/abs/2509.06172