معرفی
Ambra Demontis is an Assistant Professor at the University of Cagliari, Italy, affiliated with the Department of Electrical and Electronic Engineering. She holds an M.Sc. (Hons.) and Ph.D. in Computer Science and Electronic Engineering from the same institution. Her research focuses on enhancing the security of machine learning algorithms against adversarial attacks, including evasion, poisoning, and backdoor attacks. She actively contributes to the academic community as an Associate Editor for the International Journal of Machine Learning and Cybernetics and the Elsevier Pattern Recognition Journal, and has organized workshops like AISec (2019-2022). She is a member of IEEE, ACM, and IAPR, and chairs the IAPR Technical Committee 1.
Her educational background includes advanced degrees in Computer Science and Electronic Engineering from the University of Cagliari (M.Sc. 2014, Ph.D. 2018). She serves on conference program committees and has held roles such as Area Chair for ICPR and Track Chair for various events. Her research interests span adversarial machine learning, malware detection, and optimizing neural networks for security and energy efficiency.
Key research themes include defending against adversarial examples, improving robustness in deep learning systems, and addressing vulnerabilities in real-world applications like autonomous driving and IoT devices. She has developed frameworks such as BUILD and Attackbench for incremental learning and attack evaluation, respectively. Her work emphasizes practical defenses and theoretical insights into adversarial attack mechanisms.
Dr. Demontis’ contributions also include advancing hyperparameter optimization for adversarial attacks and exploring the intersection of domain knowledge with adversarial defense strategies. She collaborates internationally, contributing to Python tools like secml for secure machine learning. Her research addresses both theoretical and applied aspects of AI security, aiming to bridge the gap between robustness and real-world deployment challenges.
