IIT People Search

Massimiliano Pontil

Senior Researcher Tenured - Principal Investigator
Computational Statistics and Machine Learning
Phone
Research center
Biografia

Massimiliano Pontil is Senior Researcher at the Italian Institute of Technology (IIT), where he leads the Computational Statistics and Machine Learning unit, and co-director of the ELLIS Unit Genoa, a joint effort of IIT and the University of Genoa. He is also Professor of Computational Statistics and Machine Learning at University College London and member of the UCL Centre for Artificial Intelligence. He has made significant contributions to machine learning, particularly in the areas of kernel methods, multitask and transfer learning, sparsity regularization and statistical learning theory. Recent interests include meta-learning, algorithm fairness, hyperparameter optimization, and learning dynamical systems. He was awarded the Best Paper Runner Up Award from ICML 2013, an EPSRC Advanced Research Fellowship in 2006-2011, and the Edoardo R. Caianiello Award for the Best Italian PhD Thesis on Connectionism in 2002.  He has published about 180 papers at international journals and conferences. His research papers have been cited over 14,000 times and his h-index is 53 (source Scopus, July 2023). He is regularly on the programme committee of the main machine learning conferences (COLT, ICML and NeurIPS), has been on the editorial board of the Machine Learning Journal, Statistics and Computing, JMLR, and was on the Scientific Advisory Board of Max Planck Institute for Intelligent Systems during 2013-2020. He has held visiting positions at a number of universities and research institutes, including the Massachusetts Institute of Technology, the City University of Hong Kong, the Isaac Newton Institute for Mathematical Sciences in Cambridge, the University Carlos III de Madrid, ENSAE Institut Polytechnique Paris, and École Polytechnique.

Education

Title: Ph.D. in Physics
Institute: University of Genoa
Location: Genoa
Country: Italy
From: 1996 To: 1999

All Publications
2023
Grazzi R., Pontil M., Salzo S.
Bilevel Optimization with a Lower-level Contraction: Optimal Sample Complexity without Warm-Start
Journal of Machine Learning Research
Article Journal
2023
Meanti G., Chatalic A., Kostic V., Novelli P., Pontil M., Rosasco L.
Estimating Koopman operators with sketching to provably learn large scale dynamical systems
Advances in Neural Information Processing Systems
Conference Paper Conference
2023
Romeo L., Olugbade T., Pontil M., Bianchi-Berthouze N.
Multi-Rater Consensus Learning for Modeling Multiple Sparse Ratings of Affective Behaviour
IEEE Transactions on Affective Computing
Article Journal
2023
Cella L., Lounici K., Pacreau G., Pontil M.
Multi-task Representation Learning with Stochastic Linear Bandits
Proceedings of Machine Learning Research, vol. 206, pp. 4822-4847
Conference Paper Conference
2023
Ciliberto C., Pontil M., Stamos D.
Reexamining low rank matrix factorization for trace norm regularization
Mathematics In Engineering, vol. 5, (no. 3), pp. 1-22
Article Journal
Awards and Achievements
2022
Pontil M.
co-Director ELLIS Genoa Unit
2022
Pontil M.
Spoke Leader
2020
Pontil M.
Advisory Board, Max Planck Institute for Intelligent Systems, Germany
2020
Pontil M.
Elected ELLIS Fellow
2013
Pontil M.
Best Paper Runner Up
Editorships
2014-2016
Pontil M.
International Conference on Machine Learning (ICML)
2013-2023
Pontil M.
Journal of Machine Learning Research
2013-2015
Pontil M.
Neural Information Processing Systems
2009-2018
Pontil M.
Machine Learning Journal
2008-2011
Pontil M.
Annual Conference on Learning Theory (COLT)
Organized Events
2021
Chzhen E., Pontil M., Valera I.
Foundations of Algorithmic Fairness
2018
Pontil M.
American Association on Artificial Intelligence (AAAI) Area Chair
2017
Pontil M.
Workshop on Prioritising Online Content
2015
Pontil M.
Dagstuhl Workshop on "Machine Learning with Interdependent and Non-identically Distributed Data".
2012
Pontil M.
ICML Workshop entitled “Object, Functional and Structured Data: Towards Next Generation Kernel-Based Methods”