The Canadian Institute for Cybersecurity is hosting two sessions tomorrow on Deep Learning and ComputerVision.  The presenters will be Hamideh Taslimasa and Amir David, respectively. 

Each speaker will deliver their address in person and via a Microsoft Teams link, which you can access by clicking the link below.

Presenter 1: Hamideh Taslimasa
Title: Adversarial Attacks Against Network Intrusion Detection in IoT Systems
Abstract: #DeepLearning (DL) has gained popularity in network intrusion detection due to its strong capability of recognizing subtle differences between normal and malicious network activities. Although a variety of methods have been designed to leverage DL models for #security protection, whether these systems are vulnerable to adversarial examples (AEs) is unknown. In this article, the authors design a novel adversarial attack against DL-based network intrusion detection systems (NIDSs) in the Internet-of-Things environment. They successfully compromise one state-of-the-art NIDS, Kitsune.

Presenter 2: Amir David
Title: Introduction to Adversarial Attacks and Defenses in Vision Transformers
Abstract: This presentation will serve as an introduction to computer vision, particularly Vision Transformers. We will learn about Adversarial Examples in the context of #computervision and discuss why it is important to study the robustness of vision models. We will also take a look at a paper titled: “Towards Robust Vision Transformer”, in which the authors conduct a systematic robustness evaluation on the components of the Vision Transformer and leverage the robust components as building blocks for a new proposed model called Robust Vision Transformer.

Attendees can register to participate both in-person and virtually by clicking the button below to send an email to register.