Fairness in Machine Learning

Charlotte Laclau is an Associate Professor at Télécom Paris (Maître de Conférences, HDR) at Télécom Paris, Institut Polytechnique de Paris, where she leads research at the intersection of machine learning on graphs and fairness in machine learning, using optimal transport as a recurring methodological tool. She currently serves as Principal Investigator of an ANR project on fairness for graph-structured data and contributes to the national PEPR Foundry program on AI robustness and reliability, supervising several PhD students working on these questions. She is affiliated with the Hi! Paris center, the Center on Data Analytics and Artificial Intelligence for Science, Business and Society.

Antoine Gourru is an Associate Professor (Maître de Conférences) at Télécom Saint-Étienne and a member of the Machine Learning Team at the Hubert Curien Laboratory. His research focuses on responsible machine learning, with an emphasis on fairness, frugality, and explainability in natural language processing and complex data. He contributes to several funded academic projects, including ANR DIKé and ANR FAMOUS, and leads the regional i-démo DavinciDOC project on the laboratory side. He also co-supervises several PhD students and research staff working on fair NLP, frugal LLM systems, and responsible AI, and serves as co-head of the Computer Science Department at Télécom Saint-Étienne.

Winston Maxwell is Professor Emeritus at Télécom Paris – Institut Polytechnique de Paris, where he writes on subjects related to the regulation of data, AI and telecommunications. He previously had a career in private practice as a partner of the international law firm Hogan Lovells. Winston completed his law degree (JD) at Cornell, his PhD in economics at Télécom Paris, and his HDR (Habilitation à Diriger des Recherches) at the University of Paris Panthéon Sorbonne. His research focuses on the regulation of AI, and in particular human control over algorithmic systems, explainability and bias. Winston co-founded the “Operational AI Ethics” program at Telecom Paris, which includes AI Ethics teaching at Institut Polytechnique de Paris. Winston is a member of the French Academy of Technologies.

Preface.- Acknowledgements.- 1. Introduction.- 2. The Many Definitions of (Un)Fairness.- 3. Addressing Fairness.- 4. Fairness in Generative Models: Large Language Models and Image Generation.- 5. An Introduction to Fairness and the Law of Non-Discrimination.- 6.- Open Challenges.- Glossary.- Index.