Machine learning (ML) is increasingly used to implement advanced applications with nondeterministic behavior, which operate on the cloud–edge continuum. The pervasive adoption of ML is urgently calling for assurance solutions to assess applications’ nonfunctional properties (e.g., fairness, robustness, and privacy) with the aim of improving their trustworthiness. Certification has been clearly identified by policy makers, regulators, and industrial stakeholders as the preferred assurance technique to address this pressing need. Unfortunately, existing certification schemes are not immediately applicable to nondeterministic applications built on ML models. This article analyzes the challenges and deficiencies of current certification schemes, discusses open research issues, and proposes a first certification scheme for ML-based applications.

Rethinking Certification for Trustworthy Machine Learning-Based Applications / M. Anisetti, C.A. Ardagna, N. Bena, E. Damiani. - In: IEEE INTERNET COMPUTING. - ISSN 1089-7801. - 27:6(2023 Dec), pp. 22-28. [10.1109/mic.2023.3322327]

Rethinking Certification for Trustworthy Machine Learning-Based Applications

M. Anisetti
Primo
;
C.A. Ardagna
Secondo
;
N. Bena
Penultimo
;
E. Damiani
Ultimo
2023

Abstract

Machine learning (ML) is increasingly used to implement advanced applications with nondeterministic behavior, which operate on the cloud–edge continuum. The pervasive adoption of ML is urgently calling for assurance solutions to assess applications’ nonfunctional properties (e.g., fairness, robustness, and privacy) with the aim of improving their trustworthiness. Certification has been clearly identified by policy makers, regulators, and industrial stakeholders as the preferred assurance technique to address this pressing need. Unfortunately, existing certification schemes are not immediately applicable to nondeterministic applications built on ML models. This article analyzes the challenges and deficiencies of current certification schemes, discusses open research issues, and proposes a first certification scheme for ML-based applications.
Behavioral sciences; Certification; Data models; Detectors; Malware; Robustness; Security
Settore INF/01 - Informatica
   SEcurity and RIghts in the CyberSpace (SERICS)
   SERICS
   MINISTERO DELL'UNIVERSITA' E DELLA RICERCA
   codice identificativo PE00000014

   MUSA - Multilayered Urban Sustainability Actiona
   MUSA
   MINISTERO DELL'UNIVERSITA' E DELLA RICERCA
dic-2023
6-ott-2023
Article (author)
File in questo prodotto:
File Dimensione Formato  
AABD.IC2023.pdf

accesso aperto

Descrizione: Article
Tipologia: Publisher's version/PDF
Dimensione 266.88 kB
Formato Adobe PDF
266.88 kB Adobe PDF Visualizza/Apri
AABD.IC2023.pdf

accesso riservato

Descrizione: Online First
Tipologia: Post-print, accepted manuscript ecc. (versione accettata dall'editore)
Dimensione 112.08 kB
Formato Adobe PDF
112.08 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2434/1018848
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 2
  • ???jsp.display-item.citation.isi??? 0
social impact