Massi Pontil

Massi Pontil is a part-time Professor of Computational Statistics & Machine Learning in the Department of Computer Science at University College London (UCL). He joined UCL as a lecturer in 2003 and was promoted to Professor in 2010. Since 2016, his primary appointment has been at the Istituto Italiano di Tecnologia (IIT), where he leads the CSML research group. His work bridges theoretical machine learning with practical applications in physical sciences.

His research interests span a wide range of topics in machine learning theory and algorithms:

• Machine Learning Theory and Statistical Learning
• Algorithmic Fairness and Ethical AI
• Kernel Methods and Reproducing Kernel Hilbert Spaces
• Transfer Learning, Multitask Learning, and Meta-Learning
• Operator Learning and Dynamical Systems
• Sparsity Regularization and Optimization

Pontil's recent work focuses on the intersection of machine learning with numerical simulations of physical systems, particularly in molecular dynamics and climate science. His publications demonstrate a strong emphasis on theoretical foundations while addressing practical challenges in high-dimensional systems, symmetry-aware learning, and uncertainty quantification.

Among his notable honors are:

• Best Paper Runner Up Award from ICML 2013
• EPSRC Advanced Research Fellowship (2006-2011)
• Edoardo R. Caianiello Award for the Best Italian PhD Thesis on Connectionism (2002)

Professor Pontil has served on program committees for major machine learning conferences (COLT, ICML, NeurIPS) and on editorial boards of prestigious journals including Machine Learning Journal, Statistics and Computing, and JMLR. He teaches Advanced Topics in Machine Learning at UCL, with a focus on convex optimization and statistical learning theory.