Menu
Top

Program Affiliations

Biography

Professor Maurizio Filippone received his Master’s in Physics and a Ph.D. in Computer Science from the University of Genova, Italy, in 2004 and 2008, respectively. During his Ph.D. studies in 2007, Filippone spent a year as a research scholar at George Mason University, U.S.

From 2008 to 2011, he was a research associate at the University of Sheffield, U.K. (2008 to 2009), the University of Glasgow, U.K. (2010), and University College London, U.K. (2011). In 2011, Filippone took up a lecturer position at the University of Glasgow, which he left in 2015 to join EURECOM, France, as an associate professor.

In 2024, Filippone joined the Statistics program at KAUST as an associate professor.

Research Interests

Professor Filippone’s primary focus is Bayesian statistics, which enables sound decision-making through uncertainty quantification in model parameters and predictions; his main interests are in models based on deep learning and Gaussian processes.

Filippone is interested in the foundations of Bayesian statistics and computational aspects related to its use in practice. More specifically, he is developing approximations that enable recover tractability while being principled, practical and scalable.

He is also interested in applications in life and environmental sciences where uncertainty matters.

Keyword tag icon
Bayesian deep learning computational statistics Gaussian processes

Education Profile

  • Postdoc in Statistics at UCL, UK and in Computer Science at the University of Glasgow, UK, 2011-2012

  • Postdoc in Computer Science at the University of Sheffield, UK, 2009-2010

  • PhD in Computer Science, University of Genova, Italy, 2008

  • Visiting Scholar at George Mason University, Fairfax (VA), USA, 2007

  • MSc in Physics, University of Genova, Italy, 2004

Awards and Recognitions

  • Best Ph.D. Thesis Award obtained by my student Ba-Hien Tran, Doctoral School of Sorbonne University, France, 2023

  • Best Paper Award, The Biennial Pattern Recognition Journal Award, 2008

Publications

  • B.-H. Tran, G. Franzese, P. Michiardi, and M. Filippone. One-Line-of-Code Data Mollification Improves Optimization of Likelihood-based Generative Models. In NeurIPS, 2023.

  • B.-H. Tran, S. Rossi, D. Milios, and M. Filippone. All you need is a good functional prior for Bayesian deep learning. Journal of Machine Learning Research, 23(74):1--56, 2022.

  • S. Marmin and M. Filippone. Deep gaussian processes for calibration of computer models (with discussion). Bayesian Analysis, 17(4):1301-1350, 2022.

  • A. Zammit-Mangion, T. L. J. Ng, Q. Vu, and M. Filippone. Deep Compositional Spatial Models, Journal of the American Statistical Association, 117:540, 1787-1808, 2022.

  • B.-H. Tran, S. Rossi, D. Milios, P. Michiardi, E. V. Bonilla, and M. Filippone. Model selection for Bayesian autoencoders. In NeurIPS, 2021.

Research Areas

Multimedia