Archivio Istituzionale della RicercaAdversarial machine learning manipulates datasets to mislead machine learning algorithm decisions. We propose a new approach able to detect adversarial attacks, based on eXplainable and Reliable AI. The results obtained show how canonical algorithms may have difficulty in identifying attacks, while the proposed approach is able to correctly identify different adversarial settings.
On The Detection Of Adversarial Attacks Through Reliable AI / Vaccari, Ivan; Carlevaro, Alberto; Narteni, Sara; Cambiaso, Enrico; Mongelli, Maurizio. - ELETTRONICO. - (2022), pp. 1-6. (Intervento presentato al convegno IEEE INFOCOM 2022 - IEEE Conference on Computer Communications tenutosi a New York (USA) nel 02-05 May 2022) [10.1109/INFOCOMWKSHPS54753.2022.9797955].
On The Detection Of Adversarial Attacks Through Reliable AI
Narteni, Sara;
2022
Abstract
Adversarial machine learning manipulates datasets to mislead machine learning algorithm decisions. We propose a new approach able to detect adversarial attacks, based on eXplainable and Reliable AI. The results obtained show how canonical algorithms may have difficulty in identifying attacks, while the proposed approach is able to correctly identify different adversarial settings.
| File | Dimensione | Formato | |
|---|---|---|---|
|
On_The_Detection_Of_Adversarial_Attacks_Through_Reliable_AI.pdf
accesso riservato
Tipologia:
2a Post-print versione editoriale / Version of Record
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
1.64 MB
Formato
Adobe PDF
|
1.64 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
|
Infocom_Adversarial.pdf
accesso aperto
Descrizione: On the detection of adversarial attacks through Reliable AI
Tipologia:
2. Post-print / Author's Accepted Manuscript
Licenza:
Pubblico - Tutti i diritti riservati
Dimensione
1.24 MB
Formato
Adobe PDF
|
1.24 MB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2968843