PHAROS Training Series - Course 11 "Assessing and Mitigating Privacy Risks in Machine Learning and Data-Intensive Environments"

Tuesday 14 Jul 2026, 12:00 → 15:00 Europe/Athens

PHAROS AI Factory announces the 11th Course of its Training Series, under the title "Assessing and Mitigating Privacy Risks in Machine Learning and Data-Intensive Environments", under the topic AI Ethics, held online via Zoom. 

Date: July 14th, 2026, at 12:00 EEST 

Location: Online via Zoom

Presentation Language: Greek

Audience: Compliance Officers, Healthcare Experts

Prerequisites: AI Ethics

Learning Objectives: 

• Understand the privacy challenges introduced by modern data sharing and machine learning applications. 
• Learn the legal and technical differences between anonymization and pseudonymization under GDPR. 
• Become familiar with established anonymization techniques, including k-anonymity and km-anonymity, and understand their applicability and limitations. 
• Understand the principles of differential privacy and its role in privacy-preserving data analysis and machine learning. 
• Gain practical experience using the Amnesia anonymization platform. 
• Understand the most important privacy attacks targeting machine learning models and training datasets. 
• Learn about privacy-preserving machine learning approaches, including federated learning and differential privacy. 

Instructor's profile:

• Manolis Terrovitis, Athena RC

Dr. Manolis Terrovitis is a Research Director at the Information Management Systems Institute (IMSI) of the Athena Research Center in Athens, Greece. He has held several leadership positions, including serving as President of the Hellenic Accreditation System and as a member of the Board of Directors of Information Society S.A., the Greek public organization responsible for the procurement of ICT services and products for the public sector. 
He has more than 20 years of experience in data management research and in the design and development of large-scale data management systems. His primary research interests include privacy-preserving data publishing, data anonymization, and big data management. His work has been published in leading data management venues, including PVLDB, VLDB Journal, IEEE ICDE, IEEE TKDE, and others. 
Manolis has extensively worked on data anonymization algorithms and leads the development of Amnesia (https://amnesia.openaire.eu), an open-source data anonymization platform widely used by researchers and organizations for privacy-preserving data sharing.

 

Note: Please enter your institutional/corporate email when registering.

Form of Withdrawal of Consent for photos,videos.pdf

Form of Withdrawal of Consent, Registration.pdf

Read the Privacy Statement Here.pdf

12:00 → 15:00 Assessing and Mitigating Privacy Risks in Machine Learning and Data-Intensive Environments

This webinar introduces the key privacy challenges arising from machine learning and data-intensive applications. Participants will explore the principles of data anonymization, pseudonymization, and differential privacy, along with common privacy attacks targeting datasets and AI models. Through practical examples and a hands-on session with the Amnesia anonymization platform, attendees will learn how to assess and mitigate privacy risks while developing compliant and privacy-preserving AI solutions.

Speaker: Manolis Terrovitis (ATHENA RC)