Saba Heidari is a doctoral student in the Department of Computer Science studying adversarial attacks on neural networks. She participated in the spring 2024 Three Minute Thesis (3MT) Competition and qualified as one of the top 10 event finalists. The 3MT is a global research competition that challenges participants to effectively explain their research in three minutes or less to a non-specialist audience. Participants can win prizes and represent Marquette at the regional competition.
Heidari’ s 3MT presentation was titled, “Adversarial Attacks on Graph Neural Networks” and her research focuses on the field of machine learning, specifically addressing adversarial attacks as well as enhancing the robustness of neural networks, with a particular emphasis on Graph Neural Networks. The core of Heidari’ s research studies involves leveraging medical images to augment the reliability of neural networks within the medical domain.
When asked why she entered the 3MT competition, Heidari said she wanted to help raise awareness about the vulnerability of adversarial attacks. She noted that “we often hear about artificial intelligence and its usefulness in making decisions. However, it’s important to understand that AI systems can be misled by manipulated inputs, leading them to make incorrect decisions. Given that AI is increasingly being used in critical areas like self-driving cars and healthcare, it’s crucial to study how these systems can be vulnerable to such attacks. By doing so, we can work on strengthening AI systems to better withstand these challenges.”
Heidari is a research assistant in the Machine Learning, Optimization and Data Lab (MODLab) under the supervision of Dr. Nasim Yahyasoltani. She is passionate about the field of machine learning and has more than twelve journal and conference publications as well as over 50 citations recorded in her research area.
When Hedari is not engaging in research, she enjoys spending her leisure time engaging in photography, reading books and practicing yoga.
Want to read similar stories or have a success story to share? Visit Stories of Graduate Student Success to learn more.
