معرفی
Marc Juarez Miro is a Lecturer in Cyber Security and Privacy at the School of Informatics, University of Edinburgh. He is a member of the Security, Privacy, and Trust (SecPrivTru) research group, and affiliated with the Security and Privacy Research Institute and the Laboratory for Foundations of Computer Science.
His research lies at the intersection of cybersecurity, privacy, and machine learning. Key areas include ML-based traffic analysis, security and privacy of machine learning models, and fairness in private learning systems. He investigates both theoretical and practical aspects of privacy-preserving technologies, including watermarking in generative AI and defenses in anonymity networks like Tor.
Recent publications in venues such as USENIX Security, PETS, and SaTML highlight his work on vulnerabilities in watermarking schemes, fairness in differentially private ML, and traffic analysis mitigation. These works reflect a strong trend toward auditing and improving the trustworthiness of AI and privacy systems.
- Google Research Scholar Award in Security (2024)
- PETS Student Paper Award
He actively supervises PhD and Master’s students, including Kai Yao and Lisa Lavrentieva, and has secured competitive seed funding such as the GAIL grant for research on undetectable watermarks in generative AI. His advising focuses on privacy, security, and ethical AI.
Marc leads and co-organizes research initiatives such as a workshop on watermarking for generative AI and contributes to open-source tools like wfpadtools and dispa-plugin, indicating strong engagement with both theoretical and applied aspects of privacy engineering.
