Victor Manach

I am a first-year PhD student at Télécom Paris - Institut Polytechnique de Paris, within the S²A team of the LTCI laboratory. My research is supervised by Prof. Pavlo Mozharovskyi, Prof. Stephan Clémençon, and Prof. Winston Maxwell.

My research focuses on developing a statistical framework to understand and quantify bias in AI systems. I apply statistical inference methods, such as hypothesis testing and confidence intervals, to characterize deviations from fairness in machine learning models. By analyzing these deviations at both global and local levels, I aim to contribute to the design of more transparent, reliable and fair AI systems. This work is motivated by the growing need for regulated AI, specifically addressing the compliance challenges posed by the European Union's AI Act.