Othmane Marfoq - Home

I am a postdoctoral researcher at Meta in the privacy preserving machine learning team within Central Applied Science. My research interests are in machine learning, optimization, and distributed systems. Specific topics include: federated and on-device learning, large-scale machine learning and distributed optimization.

I obtained my Ph.D. from Côte d'Azur University in 2023, where I was advised by Giovanni Neglia. Prior to the PhD, I was a student at ENSTA Paris, where I graduated with a major in Applied Mathematics in 2019. In parallel of my third year at ENSTA I also graduated from the MVA Master (Mathematics, Vision, Learning) at ENS Paris-Saclay.

My research interest lies in providing theoretical understanding of distributed and federated learning systems. Inspired by the theoretical insights, I seek to design large-scale distributed/federated learning algorithms that can efficiently exploit data and system resources, with a specific attention to fairness and robustness. My research is characterized by the application of  mathematical tools from distributed optimization, information theory and statistical learning theory.