The astonishing and cryptic effectiveness of Deep Neural Networks comes with the critical vulnerability to adversarial inputs - samples maliciously crafted to confuse and hinder machine learning models. Insights into the internal representations learned by deep models can help to explain their decisions and estimate their confidence, which can enable us to trace, characterise, and filter out adversarial attacks.
Detecting adversarial inputs by looking in the black box
Carrara F;Falchi F;Amato G;
2019
Abstract
The astonishing and cryptic effectiveness of Deep Neural Networks comes with the critical vulnerability to adversarial inputs - samples maliciously crafted to confuse and hinder machine learning models. Insights into the internal representations learned by deep models can help to explain their decisions and estimate their confidence, which can enable us to trace, characterise, and filter out adversarial attacks.File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
prod_404617-doc_150368.pdf
accesso aperto
Descrizione: Detecting adversarial inputs by looking in the black box
Tipologia:
Versione Editoriale (PDF)
Dimensione
577.72 kB
Formato
Adobe PDF
|
577.72 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
