CS Seminar Explores AI and Cybersecurity Innovations

The Computer Science Department at Loyola University Chicago recently hosted a CS Seminar Series on Friday, November 22nd, featuring insightful talks by Dr. Daniel Moreira and Mr. Josh Honig alongside Mr. Nathan Ferrell, all from Loyola University Chicago. The event was held in the Regis Hall MPR Room and offered attendees an engaging afternoon filled with thought-provoking discussions, networking opportunities, and a shared meal.

Event Schedule:

• Location: Regis Hall - MPR Room 132-133 (6340 N. Winthrop Ave.)
• 12:00 - 12:40 PM: Lunch and Social
• 12:40 - 01:20 PM: Talk by Dr. Daniel Moreira (Loyola University Chicago)
• 01:20 - 02:00 PM: Talk by Mr. Josh Honig and Mr. Nathan Ferrell (Loyola University Chicago)

Talks and Speakers:

1. Explainable Artifacts for Synthetic Western Blot Source Attribution

Talk by Dr. Daniel Moreira (Loyola University Chicago)

Dr. Daniel Moreira delved into recent advancements in Artificial Intelligence that enable the creation of synthetic scientific images indistinguishable from authentic ones. He highlighted the challenges posed by "paper mills," organizations that systematically generate fraudulent articles using these technologies, thereby undermining trust in scientific research. Dr. Moreira presented his latest efforts to identify explainable artifacts generated by state-of-the-art generative models like Generative Adversarial Networks and Diffusion Models. His work focuses on leveraging these artifacts for open-set identification and source attribution, effectively pinpointing the models that created the synthetic images.

Speaker Bio:

Dr. Daniel Moreira received his Ph.D. in Computer Science from the University of Campinas, Brazil, in 2016. After four years as a systems analyst with the Brazilian Federal Data Processing Service (SERPRO), he joined the University of Notre Dame as a post-doctoral fellow and later as an assistant research professor. Currently, he is an assistant professor in the Department of Computer Science at Loyola University Chicago. Dr. Moreira also serves as an associate editor for IEEE Transactions on Information Forensics and Security (T-IFS) and Elsevier's Pattern Recognition journals. His research interests include media forensics, machine learning, computer vision, and biometrics.

2. Website Fingerprinting: Using ML to Predict User Behavior from Encrypted Metadata

Talk by Mr. Josh Honig and Mr. Nathan Ferrell (Loyola University Chicago)

Mr. Josh Honig and Mr. Nathan Ferrell presented their research on how Machine Learning can be utilized to identify user web browsing behavior based solely on the metadata of encrypted network traffic. They demonstrated that with just the size, direction, and order of encrypted HTTPS packets, they could predict which website a user was visiting using a Random Forest classifier—with reasonable accuracy. The described threat model is concerningly easy for a lone attacker to establish, requiring minimal visibility and computational resources.

Speaker Bios:

Mr. Josh Honig is a graduate student at Loyola University Chicago, having completed his undergraduate studies there as well. Engaged in research on Website Fingerprinting techniques with Dr. Chan-Tin since 2022, Josh enjoys participating in cybersecurity competitions and serves as the president of Loyola's cybersecurity competition club, 7968.

Mr. Nathan Ferrell is an undergraduate student pursuing a Bachelor's degree in Cybersecurity at Loyola University Chicago. Outside of his courses, he is involved in undergraduate research and actively participates in numerous cybersecurity competitions.

The Computer Science Department invited students, faculty, and the broader community to attend this seminar to gain deeper insights into the challenges and advancements in artificial intelligence, media forensics, and cybersecurity.

Note of Appreciation: The Computer Science Department would like to extend its gratitude to our esteemed speakers—Dr. Daniel Moreira, Mr. Josh Honig, and Mr. Nathan Ferrell—for sharing their knowledge and research findings. We also thank all the faculty, staff, and students who contributed their time and effort to make this event a success.

The department looks forward to hosting more events that contribute to the academic and professional growth of its community.