AI for Security and Security of AI

WORKSHOP

Mon June 20, 2022

Overview

Recent years have seen a dramatic increase in applications of Machine Learning (ML) and Artificial Intelligence (AI) to security and privacy problems. Such widespread adoption fosters new lines of research, where ML and AI techniques help fight threats side to side with already-deployed security solutions. However, the same abundance of AI and ML techniques also raises doubts about their intrinsic security, as they might be the target of the next cyber-attacks that leverage their weak spot, becoming de-facto the weakest link in the security chain to attack.

To address these fundamental issues, this workshop focuses on inspiring and leading technical discussions around the entanglement between cyber security and ML/AI. In particular, our interests are twofold: (i) AI for security, referring to the analysis and study of all the cyber-security applications that can be improved and automated by ML/AI techniques (like malware, spam, phishing, and botnet detection); (ii) security of AI, since we are interested in understanding how ML/AI technologies can be spoiled by skilled attackers (e.g., data poisoning, adversarial examples, and privacy-related threats), and how to develop novel techniques to harden existing ML/AI solutions (e.g. robust training, data augmentation, domain knowledge). 

By investigating these aspects, we believe we will spectate the development of the next generation of ML/AI applications. These will not only improve the security of other technologies, by making them more effective in detecting threats, but also they will stand as trustworthy techniques that can face skilled adversaries in the wild without being subdued.

Topics of Interest

Topics of interest include (but are not limited to):

AI for Security

  • Spam and malware detection
  • Phishing detection and prevention
  • Botnet detection
  • Intrusion detection and response
  • Security in social networks
  • Biometric identification/verification
  • User authentication

Security of AI

  • Adversarial machine learning
  • Security of deep learning systems
  • Attacks and defenses on machine learning and AI (e.g., adversarial examples, data poisoning, privacy-related attacks)
  • Robust statistics
  • Privacy-preserving machine learning

Program

h. 14:00 – 14:20
Welcome and Event Presentation

h. 14:20 – 15:20
Invited Talk: TBD

h. 15:20 – 15:40
C-OSINT: COVID-19 Open Source artificial INTelligence framework
Leonardo RANALDI (presenter)

Department of Innovation and Information Engineering, Guglielmo Marconi University, Roma, Italy
Aria NOURBAKHSH
Department of Enterprise Engineering, University of Rome Tor Vergata, Roma, Italy
Francesca FALLUCCHI
Department of Innovation and Information Engineering, Guglielmo Marconi University, Roma, Italy
Fabio Massimo ZANZOTTO
Department of Enterprise Engineering, University of Rome Tor Vergata, Roma, Italy

h. 15:40 – 16:00
Improving Malware Detection with Explainable Machine Learning
Michele SCALAS

Konrad RIECK
TU Braunschweig

Giorgio GIACINTO (presenter)
University of Cagliari

h. 16:00 – 16:30
Coffee break

h. 16:30 – 16:50
Machine Learning Techniques for Italian Phishing Detection
Leonardo RANALDI (presenter)

Department of Innovation and Information Engineering, Guglielmo Marconi University, Roma, Italy
Michele PETITO
Agency for Digital Italy (AgID), Roma, Italy
Marco GERARDI
Department of Innovation and Information Engineering
Guglielmo MARCONI
University, Roma, Italy
Francesca FALLUCCHI
Department of Innovation and Information Engineering, Guglielmo Marconi University, Roma, Italy
Fabio Massimo ZANZOTTO
Department of Enterprise Engineering, University of Rome Tor Vergata, Roma, Italy

h. 16:50 – 17:10
Evasion Attacks against Banking Fraud Detection Systems
Michele CARMINATI (presenter)

Politecnico di Milano
Luca SANTINI
Politecnico di Milano
Mario POLINO
Politecnico di Milano
Stefano ZANERO
Politecnico di Milano

h. 17:10 – 17:30
Energy-Latency Attacks via Sponge Poisoning
Antonio Emanuele CINÀ (presenter)

Ca’ Foscari University of Venice
Ambra DEMONTIS
University of Cagliari
Battista BIGGIO
University of Cagliari, Pluribus One
Fabio ROLI
University of Genova, Pluribus One
Marcello PELILLO
Ca’ Foscari University of Venice

h. 17:30 – 17:50
Resilience Verification of Tree-Based Classifiers
Stefano CALZAVARA

Ca’ Foscari University of Venice
Lorenzo CAZZARO (presenter)
Ca’ Foscari University of Venice
Claudio LUCCHESE
Ca’ Foscari University of Venice
Federico MARCUZZI
Ca’ Foscari University of Venice
Salvatore ORLANDO
Ca’ Foscari University of Venice

h. 17:50 – 18:00
Concluding Remarks

Committee

Workshop Chairs

  • Battista BIGGIO
    Assistant Professor, University of Cagliari, Italy; Pluribus One
  • Maura PINTOR
    Postdoctoral Researcher, University of Cagliari, Italy; Pluribus One
  • Luca DEMETRIO
    Postdoctoral Researcher, University of Cagliari, Italy; Pluribus One
  • Fabio ROLI
    Professor, University of Genova, Italy; Pluribus One

Call

Submission Guidelines

Papers must be in English, formatted in pdf according to the ITASEC conference template (Easychair style: https://easychair.org/publications/for_authors) and no longer than 10 pages, excluding bibliography. This workshop has no official proceedings, so we will also accept submissions that have been published elsewhere, provided that this is clearly acknowledged in the submission (e.g., with a footnote on the first page reporting the full reference), and that the submission is adapted according to the given template and page limits.

Submission Site

Submission link: https://easychair.org/cfp/aissai22

Important Dates
  • May 1 (extend to 15th) 2022: Workshop submission deadline
  • May 31, 2022: Workshop paper acceptance results
  • June 10, 2022: Workshop camera-ready version
  • June 20, 2022 Workshop day