Workshop Program
Where: UNIBO Aula Studio / Plesso Belmeloro: Room Q
When: Monday, February 3rd, 14:30-18:30
Schedule
14:30-14:45
Workshop Opening and Greetings
14:45-15:40
Invited Guest Talk
Speaker: Giuseppina Andresini
Title: “Explainability to Foster Transparency and Trust in AI-powered Cybersecurity”
Abstract: Considering the acceleration of digital transformation, the need to protect private and public organisations from cyber-threats has never been more critical. Artificial Intelligence (AI), particularly Deep Learning, is proven to be very effective in mitigating cyber-threat risks caused by network intrusions, malware or spam. However, Deep Learning techniques train accurate classification models that are typically “black boxes”, while explainable AI decisions can increase security stakeholder confidence. On the other hand, the emerging Symbiotic AI paradigm is contributing to the development of eXplainable AI (XAI) systems whose decisions are achieved in symbiosis with humans to mitigate errors, identify biases, and ensure fairness and transparency. The talk will provide an exploration of XAI methods recently used in cybersecurity literature to foster transparency and trust in AI-powered cybersecurity systems by improving the reliability and robustness of AI models, especially in scenarios involving imbalanced data, concept drift, and adversarial attacks.
15:40-16:00
Paper Presentation
Title: “Deep Generative Models As An Adversarial Attack Strategy For Tabular Machine Learning”
Authors: Salijona Dyrmishi, Mihaela Cătălina Stoian, Eleonora Giunchiglia, and Maxime Cordy
Presented by: Maxime Cordy
16:00-16:15
Break
16:15-16:35
Paper Presentation
Title: “RobustBlack: Challenging Black-Box Adversarial Attacks on State-of-the-Art Defenses”
Authors: Mohamed Djilani, Salah Ghamizi, and Maxime Cordy
Presented by: Mohamed Djilani
16:35-16:55
Paper Presentation
Title: “Verifiable Boosted Tree Ensembles”
Authors: Stefano Calzavara, Lorenzo Cazzaro, Claudio Lucchese, and Giulio Ermanno Pibiri
Presented by: Lorenzo Cazzaro
16:55-17:15
Paper Presentation
Title: “TabularBench: Benchmarking Adversarial Robustness for Tabular Deep Learning in Real-world Use-cases”
Authors: Thibault Simonetto, Salah Ghamizi, and Maxime Cordy
Presented by: Maxime Cordy
17:15-17:35
Invited Student Talk
Speaker: Maria Camporese
Title: “Using ML filters to help automated vulnerability repairs: when it helps and when it doesn’t“
17:35-17:45
Workshop Closing Remarks