Christos, pursuing a PhD in explainable AI and fairness in networks at the University of Ioannina, holds an MEng with specialization in Artificial Intelligence and Data Science from the same university.

He has research and work experience in machine learning, data mining, natural language processing and graph analysis. His thesis was on the application of machine learning techniques for creating embeddings for relational data. From November 2022 to October 2023, he worked as a research associate at university of Ioannina, collaborating with the CDI of Pfizer in Thessaloniki where he focused on developing innovative solutions for addressing healthcare misinformation detection on social media platforms.

RESEARCH INTERESTS

Christos’ research interests include Explainable AI and fairness in networks. Specifically, his research will focus on areas such as counterfactual explanations, fairness enhancement for individuals and groups, graph-based algorithm explanations, and exploring contrastive explanations for specific input data pairs.

PhD research on "Counterfactual Explanations"
Abstract: AI makes decisions in loans and jobs. Models can inherit or introduce bias, so we audit group fairness using group counterfactual explanations. Our approach FACEGroup builds a feasibility graph, partitions groups into subgroups, and optimizes cost/coverage trade-offs (MIP/Greedy). New metrics reveal subgroup disparities, while FACEGroup yields feasible, compact counterfactual sets.