Certifying machine learning systems has become more and more important, especially when they are deployed in safety critical domains. In this paper, we sum up the work in [1] that combines Deep Learning with Formal Methods for the automated synthesis of certified neural networks and we discuss current open research lines.
Automated Synthesis of Certified Neural Networks: Initial Results and Open Research Lines
Matteo Zavatteri
Writing – Original Draft Preparation
;Davide BresolinWriting – Review & Editing
;Nicolo NavarinWriting – Review & Editing
2025
Abstract
Certifying machine learning systems has become more and more important, especially when they are deployed in safety critical domains. In this paper, we sum up the work in [1] that combines Deep Learning with Formal Methods for the automated synthesis of certified neural networks and we discuss current open research lines.
File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
OVERLAY-2024-certified-NN.pdf
accesso aperto
Tipologia:
Published (Publisher's Version of Record)
Licenza:
Creative commons
Dimensione
986.94 kB
Formato
Adobe PDF
|
986.94 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
