|
Time |
Title |
Presenter/Author |
|
14:00 –
14:05 |
Opening |
Cristian Bucur |
|
14:05 –
14:30 |
IoT
Attack-Traffic Classification using Deep Learning: A First Look at Explainability and Robustness |
Antonio Montieri |
|
14:30 –
14:55 |
A Max-Min Security Game for Coordinated Backdoor Attacks on
Federated Learning |
Omar Abdel Wahab and Anderson Avila |
|
14:55 –
15:20 |
One to rule them all: A study of requirement management
tools supporting the development of modern AI-based software |
Abdul-Rasheed Ottun, Mehrdad
Asadi, Michell Boerger, Nikolay Tcholtchev, Joao
Fernando Ferreira Goncalves, Dušan Borovčanin, Bartlomiej Siniarski, and Huber Flores |
|
|
Coffee Break |
|
|
15:35 –
16:00 |
Adaptive Intrusion Detection Systems: Class Incremental
Learning for IoT Emerging Threats |
Francesco Cerasuolo, Giampaolo Bovenzi, Christian Marescalco, Francesco Cirillo, Domenico Ciuonzo, and Antonio Pescapè |
|
16:00 –
16:25 |
A Novel Approach to Parallel Anomaly Detection: Application
in Cybersecurity |
Zineb Ziani, Nahid Emad, and Ahmed Bouaziz |
|
16:25 –
16:30 |
Closing remarks |
Cristian Bucur |
Today, Cyber intrusions occur at the speed of AI. Both attackers and defenders use newly developed techniques in AI, and ML. IoT systems are especially vulnerable and intrusion information and monitoring is beyond the capabilities of the human workforce. We must automate intrusion detection and monitoring of IoT to keep up with BigData, generated by activities of IoTs. We propose to invite research presentations from both academia and industry in the area of security for IoTs with a focus on the use of Machine Learning and AI.
Securing IoTs is a very specific branch of cybersecurity that requires expertise in IoTs limited operating systems and network capabilities. General cybersecurity personnel monitoring organizations in their Security Operation Centers do not have the necessary knowledge and skill to recognize IoT attacks. IoT specialists are rare. Using AI and ML, automation of intrusion and monitoring of IoTs can be automated. We propose to bring together researchers in this specific area of cybersecurity to share ideas and collaborate, thus creating a community of research and practice.
cristian.bucur@polymtl.ca
bucurc@gmail.com