معرفی
Frank NIELSEN is a Professor at École Polytechnique with expertise in information geometry, data science, and machine learning. He holds a PhD (1996) and HDR (2006) in computer science and has established himself as a leading researcher in geometric approaches to information science.
His educational background includes a PhD in computer science (1996) followed by a Habilitation à Diriger des Recherches (HDR) in 2006, the highest academic qualification in France that qualifies one to supervise doctoral candidates.
Dr. NIELSEN's research focuses on the Geometric Science of Information, where he develops theoretical frameworks for understanding data through geometric and information-theoretic lenses. His work bridges
- Computational information geometry
- Statistical manifold theory
- Bregman divergences and their applications
- Machine learning with geometric foundations
- High-dimensional data analysis
His extensive publication record shows a clear trend toward developing geometric frameworks for understanding statistical divergences, with recent work focusing on Bregman geometry, Fisher-Rao metrics, and their applications in machine learning. His research spans theoretical developments in information geometry to practical implementations like the pyBregMan Python library, demonstrating both theoretical depth and practical relevance.
Dr. NIELSEN has made significant contributions through his teaching and publications. He has taught courses at École Polytechnique including INF442, INF517, and INF591. His authored textbooks include Introduction to HPC with MPI for Data Science (2016), A Concise and Practical Introduction to Programming Algorithms in Java (2009), and Visual Computing: Geometry, Graphics, and Vision (2005). He has also edited influential volumes such as Computational Information Geometry for Image and Signal Processing (2016) and Geometric Theory of Information (2014).
He actively organizes and participates in academic events, serving on program committees for major conferences including GSI (Geometric Science of Information), CVPR, and ICCV. His work has established him as a key figure in the growing field of geometric approaches to information science.
