Piotr Zwiernik

Piotr Zwiernik is a Professor Agregat in the Department of Economics and Business at Universitat Pompeu Fabra (UPF), Barcelona, and a member of the BSE Data Science Center. He is currently on leave from the Department of Statistical Sciences and the Department of Mathematics at the University of Toronto. His research bridges statistics, algebraic geometry, and machine learning, focusing on graphical models, covariance estimation, tensors, and algebraic methods in statistics.

Research Interests:

• Graphical Models
• Covariance Matrix Estimation
• Convex Analysis
• Tensors
• Algebraic and Combinatorial Methods in Statistics
• Statistical Learning Theory

His recent publications span high-impact journals such as the Annals of Statistics, Biometrika, and Journal of the Royal Statistical Society. The work demonstrates a consistent trend in developing mathematically rigorous frameworks for understanding dependence structures, latent variable models, and high-dimensional inference, often leveraging tools from algebraic geometry and convex optimization. Key themes include total positivity, tensor methods, and the geometry of statistical models.

Scientific Recognition:

• Editorial Board Member, Journal of the Royal Statistical Society Series B (JRSS-B)
• Editorial Board Member, Biometrika
• Editorial Board Member, Scandinavian Journal of Statistics
• Editorial Board Member, Algebraic Statistics

Zwiernik actively mentors students and is seeking PhD candidates with strong mathematical backgrounds at UPF and the Institute of Mathematics of UPC. He has led the development of several open-source R packages, including golazo, MTP2binary, and StructuralEM, facilitating research in graphical models and latent structures. His work is supported by a network of collaborations with leading statisticians and has significant theoretical and applied implications in data science.

Laboratories and Research Groups:

• Statistics@UPF
• BSE Data Science Center