Prof. Dr Kassem Kallas, Inserm, France
Dr. K. Kallas is a multidisciplinary Research Scientist and Junior Professor specializing in Artificial Intelligence (AI) ๐ค and Cybersecurity ๐. With a Ph.D. in Information Engineering from the University of Siena ๐ฎ๐น, he is currently a Senior Scientist at the French National Institute of Health and Medical Research ๐ซ๐ท. He has held prestigious roles at INRIA and the U.S. National Institute of Standards and Technology (NIST) ๐บ๐ธ. Dr. Kallas is known for pioneering research in adversarial deep learning, game-theoretic sensor fusion, and AI intellectual property protection via watermarking. A recognized speaker ๐ค and mentor, he actively contributes to global academic and industry collaborations. He also volunteers with the IEEE Collabratec and Lebanese Red Cross โค๏ธ. He is pursuing the Habilitation ร Diriger des Recherches (HDR), the highest academic qualification in France. His work bridges the worlds of AI security, strategic leadership, and ethical innovation in digital technologies.
Publication Profile
orcid
Education
Dr. Kallas earned his Ph.D. in Information Engineering and Sciences (2013โ2017) from the University of Siena, Italy ๐ฎ๐น, with a dissertation on game-theoretic approaches to adversarial information fusion. He is currently pursuing the Habilitation ร Diriger des Recherches (HDR) at the University of Western Brittany ๐ซ๐ท (2024โ2025), focusing on AI security through backdoor attack analysis and watermarking. He also holds an Executive MBA ๐ in Strategic Leadership from Valar Institute, Quantic School of Business and Technology (2023โ2024), graduating with a stellar 94.8% average. Earlier, he completed a Second Level Master in Wireless Systems ๐ก at Politecnico di Torino ๐ฎ๐น (2012โ2013), an M.Sc. in Computer and Communications Engineering from the Lebanese International University ๐ฑ๐ง (2010โ2012), and a B.Sc. in Telecommunications Engineering (2006โ2010) from the same institution. His academic path blends engineering, leadership, and innovation at the highest international levels ๐.
Experience
Dr. Kallas is currently a Senior Scientist at the French National Institute of Health and Medical Research ๐งฌ, where he leads research on secure and private AI in healthcare. From 2022โ2023, he served as a Research Scientist at INRIA ๐ซ๐ท, contributing to the SAIDA defense AI security project, with focus areas including backdoor attacks, model defenses, and neural watermarking. Previously, he was a Research Fellow at NIST ๐บ๐ธ (2020โ2022), working in the wireless communications division of the chemical and nuclear measurement group. His diverse career includes involvement in DARPA, the U.S. Air Force Research Lab, French ANR, and the Italian Ministry of Research. As a speaker and academic contributor, he has presented globally ๐ on AI threats and defenses, quantum neural networks, and adversarial machine learning. He is a mentor at IEEE Collabratec and a youth volunteer with the Lebanese Red Cross ๐, blending scientific leadership with social responsibility.
Awards and Honors
Dr. Kallas has received numerous awards and recognitions across his career. His Ph.D. thesis was ranked in the Top 3 Best-of-the-Best by Springer ๐ฅ. He earned the Best Paper Award ๐
at the 9th International Conference on Advances in Multimedia (MMEDIA 2017), and his ICASSP 2023 paper was ranked in the Top 3% ๐ฅ for its groundbreaking contributions to DNN watermarking. He was selected as an Invited Keynote Speaker ๐๏ธ at international conferences, including the 5th ICCCS in India, TheAIEngineers in Lebanon, and seminars at IMT Atlantique and รcole Polytechnique in France. His work is regularly featured in high-impact publications and global research events. Beyond academia, his leadership was recognized in his EMBA program, where he led a business consultancy team to full marks โญ. These accolades reflect his innovation, influence, and impact across cybersecurity, AI, and signal processing
Research Focus
Dr. Kallasโs research focuses on AI security, adversarial machine learning, and cybersecurity for distributed systems ๐. His pioneering work investigates backdoor attacks, model robustness, and the protection of AI intellectual property via watermarking ๐ง. Through a game-theoretic lens, he analyzes adversarial behavior in sensor networks, making his research crucial for defense, healthcare, and IoT systems. At INRIA, he contributed to SAIDA, a project focused on securing deep learning systems against hidden threats. His current role at INSERM emphasizes the privacy-preserving deployment of AI models in healthcare, addressing critical issues in ethical AI. He also explores quantum neural networks, signal processing, and secure fusion techniques, enabling more resilient AI ecosystems. With involvement in EU and US-funded defense and science initiatives (DARPA, ANR, etc.), his interdisciplinary approach bridges theoretical foundations with practical solutions. His aim: building robust, transparent, and accountable AI systems fit for complex, real-world deployments ๐.
Publication Top Notes
-
๐ Deciphering the Realm of Artificial Intelligence Security: Journeying from Backdoor Attacks in Deep Learning to Safeguarding Their Intellectual Property Through Watermarking (HDR Dissertation, 2025)
-
๐ Simplifying Care, Amplifying Impact: ADDYOU โ Your Path to Well-Being (EMBA Final Project, 2024)
-
๐ A Game-Theoretic Approach for Adversarial Information Fusion in Distributed Sensor Networks (PhD Dissertation, 2017)
-
๐ Design of Capacity Control for TCP Protocol using Markov Chains (Master Thesis, Politecnico di Torino)
-
๐ Simulation of Bit-Interleaved LDPC with Iterative Decoding System (M.Sc. Thesis)
-
๐ Design and Hardware Implementation of Wireless Liquid Level Indicator System (B.Sc. Final Project)