DESCRIPTION
AI bias is an emerging concern in the field of AI, due to the widespread deployment of AI-based services and applications with ubiquitous effects on our daily lives. AI bias is particularly important in high-stakes decision making scenarios such as CV ranking and hiring, credit scoring and recidivism prediction, but it is also becoming a growing concern in the context of generative AI systems, where harmful stereotypes are perpetuated and amplified through modern foundational This tutorial will introduce the emerging field of AI bias and fairness, starting from the general AI setting, and then will proceed with a more in-depth study of the problem in the context of computer vision models and applications. It will also include two hands- on interactive sessions, where participants will have the opportunity to experiment with methods for assessing and mitigating bias, first in general tabular datasets and then in computer vision datasets and models.
DETAILS
Course type: Tutorial
Duration (tentative):
Introduction to AI bias and fairness (E. Ntoutsi) – 45 min – 1 hour
Bias in computer vision (S. Papadopoulos) – 45 min – 1 hour
Overview of visual bias mitigation approaches (C. Diou) – 45 min – 1 hour Hands-on session on FairBench for assessing bias (E. Krasanakis) – 1:30-2 hours
Hands-on session on bias assessment/mitigation in computer vision (G. Sarridis) – 1:30-2 hours
Institutions of lecturers: Centre for Research and Technology Hellas (CERTH), Harokopio University Athens (HUA), University of the Bundeswehr Munich (UniBwΜ).
LECTURERS
Dr. Symeon Papadopoulos is a Principal Researcher at the Information Technologies Institute, Centre for Research and Technology Hellas (CERTH-ITI), Thessaloniki, Greece. He holds a PhD degree in Computer Science from the Aristotle University of Thessaloniki (2012) on the topic of Knowledge discovery from large-scale mining of social media content. His research interests lie at the intersection of multimedia understanding, social network analysis, information retrieval, big data management and artificial intelligence. Dr. Papadopoulos has co-authored more than 50 papers in refereed journals, 10 book chapters and 140 papers in conferences, 3 patents/workshops, and has co-edited two books. He participates in and coordinates a number of relevant EC FP7, H2020 and HE projects in media convergence, social media and artificial intelligence. He is heading the Media Analysis, Verification and Retrieval Group (MeVer), and is a co-founder of the Infalia Private Company, a spin-out of CERTH-ITI.
Prof. Dr. Christos Diou received the Diploma in Electrical and Computer Engineering and the Ph.D. degree on the analysis of multimedia with machine learning from the Aristotle University of Thessaloniki, Thessaloniki, Greece. He is currently an Associate Professor of Artificial Intelligence and Machine Learning with the Department of Informatics and Telematics, Harokopio University of Athens. His main research interests include robust representation learning capable of out-of-distribution generalization, methods for fair and interpretable machine learning, as well as the use of machine learning for the estimation of causal effects from observational data. He has published over 100 articles in international scientific journals and conferences and has more than 15 years of experience participating and leading European and national research projects, focusing on applications of artificial intelligence in healthcare.
Ioannis Sarridis received both his bachelor’s and master’s degrees from the Department of Informatics at Aristotle University of Thessaloniki, Thessaloniki, Greece, in 2018 and 2020, respectively. He is currently pursuing a Ph.D. degree at Harokopio University of Athens and is a Research Associate with the Centre for Research and Technology Hellas, Thessaloniki, Greece. His research interests include AI fairness, computer vision, deep learning, and hypergraph learning.
Dr. Emmanouil Krasanakis is a PostDoc Researcher at the Information Technologies Institute, Centre for Research and Technology Hellas (CERTH-ITI). He holds a PhD degree in Computer Science from the Aristotle University of Thessaloniki (2022) on the topic of algorithmic-aided software development in new domains, and a BSc degree from the same faculty. His main research interests include graph theory and graph neural networks, algebraic methods, machine learning with focus on algorithmic fairness and discrimination, and software engineering. Dr. Krasanakis has co-authored 9 papers in refereed journals and 12 conference papers. He has participated in a number of relevant H2020 projects, out of which he is serving as the technical lead of MAMMOth.
Prof. Dr. Eirini Ntoutsi is a full professor at the University of the Bundeswehr (UniBwM) in Munich, Germany. Before joining UniBwM, she was a full professor at the Free University Berlin, Germany and an associate professor at the Leibniz University of Hannover, Germany. She completed her postdoc at LMU Munich and obtained her Ph.D. (2008) from the University of Piraeus, Athens, Greece, on the topic of machine learning model comparison. She holds a master’s (2003) and diploma (2001) in Computer Engineering and Informatics from the University of Patras, Greece. Her research is in the area of AI/ML with the aim of creating intelligent systems that benefit society. Her work addresses real-world data challenges, such as imbalances and non-stationarity, and the responsible development of AI by tackling bias and discrimination, improving transparency through XAI, and strengthening the robustness of AI/ML models. She is an active member of the research community, recently serving as PC co-chair for the research track of ECML/PKDD 2024.