Webinar: Preserving Data Correlations in Adversarial Network Traffic

#Adversarial #Network #Traffic #deep-learning #intrusion-detection #IEEE #savannah #detection-systems
Share
Title: Preserving Data Correlations in Adversarial Network Traffic
 
Abstract: Intrusion Detection Systems (IDS) based on machine learning (ML)/deep learning (DL) have been developed to detect malicious activities within communication networks. However, existing ML/DL-based IDS are vulnerable to adversarial traffic, which is generated by deliberately adding perturbations to normal traffic in order to maximize classification error. Most of the existing adversarial traffic generation methods overlook the correlations between data features. Failing to account for these correlations results in unrealistic adversarial samples that deviate from the original data distribution. In this talk, we present Constraint-based Adversarial traffic Generation (CAG), a novel scheme that is capable of evading ML/DL-based IDS while preserving data correlations. Compared to the existing mechanisms, CAG successfully takes the linear and zero multiplication correlations involving three features into consideration. Our experimental results indicate that CAG can evade ML/DL-based IDS in both white-box and black-box attack scenarios.
 
Speaker Bio: Qiang Ye (https://www.dal.ca/faculty/computerscience/faculty-staff/qiang-ye.html) is a Professor in the Faculty of Computer Science at Dalhousie University, Canada. His current research interests lie in the area of communication networks in general. Specifically, he is interested in Wireless Networks, Internet of Things, Network Security, and Machine Learning. He has published a series of papers in top publication venues such as IEEE/ACM Transactions on Networking (TON), IEEE Transactions on Parallel and Distributed Systems (TPDS), IEEE Transactions on Wireless Communications (TWC), IEEE International Conference on Computer Communications (INFOCOM), and ACM International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc). He has been the co-symposium chair for several tier-1 international conferences such as IEEE Global Communications Conference (GLOBECOM). He received a Ph.D. in Computing Science from the University of Alberta in 2007. His M. Engr. and B. Engr. in Computer Science and Technology are from Harbin Institute of Technology, P.R. China. He is a Senior Member of IEEE.
 
To Register: (To receive Zoom Link: sign up at https://forms.gle/deTHof3JtJ8NLAH56) 
 

  Date and Time

  Location

  Hosts

  Registration


  • Add_To_Calendar_icon Add Event to Calendar
If you are not a robot, please complete the ReCAPTCHA to display virtual attendance info.
  • Contact Event Hosts
  • Starts 12 June 2025 04:00 AM UTC
  • Ends 19 June 2025 04:00 AM UTC
  • No Admission Charge

  Speakers

Professor Qiang Ye of Dalhousie University

Topic:

Title: Preserving Data Correlations in Adversarial Network Traffic

Abstract: Intrusion Detection Systems (IDS) based on machine learning (ML)/deep learning (DL) have been developed to detect malicious activities within communication networks. However, existing ML/DL-based IDS are vulnerable to adversarial traffic, which is generated by deliberately adding perturbations to normal traffic in order to maximize classification error. Most of the existing adversarial traffic generation methods overlook the correlations between data features. Failing to account for these correlations results in unrealistic adversarial samples that deviate from the original data distribution. In this talk, we present Constraint-based Adversarial traffic Generation (CAG), a novel scheme that is capable of evading ML/DL-based IDS while preserving data correlations. Compared to the existing mechanisms, CAG successfully takes the linear and zero multiplication correlations involving three features into consideration. Our experimental results indicate that CAG can evade ML/DL-based IDS in both white-box and black-box attack scenarios.

Biography:

Bio: Qiang Ye (https://www.dal.ca/faculty/computerscience/faculty-staff/qiang-ye.html) is a Professor in the Faculty of Computer Science at Dalhousie University, Canada. His current research interests lie in the area of communication networks in general. Specifically, he is interested in Wireless Networks, Internet of Things, Network Security, and Machine Learning. He has published a series of papers in top publication venues such as IEEE/ACM Transactions on Networking (TON), IEEE Transactions on Parallel and Distributed Systems (TPDS), IEEE Transactions on Wireless Communications (TWC), IEEE International Conference on Computer Communications (INFOCOM), and ACM International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc). He has been the co-symposium chair for several tier-1 international conferences such as IEEE Global Communications Conference (GLOBECOM). He received a Ph.D. in Computing Science from the University of Alberta in 2007. His M. Engr. and B. Engr. in Computer Science and Technology are from Harbin Institute of Technology, P.R. China. He is a Senior Member of IEEE.