Raffaele Marino

Scientist
raffaele.marino@epfl.ch +41 21 693 81 11
EPFL IC IINFCOM LTHC
INR 139 (Bâtiment INR)
Station 14
CH-1015 Lausanne
Web site: Web site: https://lthcwww.epfl.ch/
Biography
I am a theoretical physicist with a background on the theory of transport processes in non-equilibrium systems, where thermal noise typically plays a dominant role. I worked at The Hebrew University of Jerusalem under the supervision of Scott Kirkpatrick on stochastic optimisation, computational complexity, and graph theory. The aim of my research was developing new greedy algorithms and message passing algorithms for probabilistic graphical models. Today I am working at EPFL, in collaboration with Nicolas Macris, on fields of high dimensional statistics, modern inference, and machine learning. My position is funded by the ANR-FNS French-Swiss grant PAIL (Phase diagrams and Algorithms for Inference and Learning).Current work
My research is focused on fields of high dimensional statistics, modern inference, and machine learning, and is funded by the ANR-FNS French-Swiss grant PAIL (Phase diagrams and Algorithms for Inference and Learning).The aim of the present interdisciplinary project is to benefit from the recent developments in fields as diverse as theoretical statistical physics coding, signal processing, probability and random combinatorial optimization to uncover principles hidden behind the success of the new approaches in modern inference problems.
Research goals are:
1. Precisely formulate the models of single level and multilevel inference systems in the framework of probabilistic graphical models. Identify the relevant asymptotic limits and construct or generate relevant priors capturing the notion of feature are among our objectives.
2. Thoroughly analyze the phase diagrams of such systems by the methods of statistical physics of spin glasses. Use this analysis to obtain new algorithms and determine fundamental limits to learning features. This includes both information theoretic aspects (the “static” behavior in physics) and algorithmic ones (where the “dynamics” of the algorithm is studied with statistical physics methods)
3. Use the physics predictions to develop a mathematically coherent theory of the algorithmic and optimal phase transitions. Leverage on recent mathematical methods developed in the last few years in the context of coding, signal processing, and constraint satisfaction.
Education
Postdoctoral Research Fellow
Computer Science and Theoretical Physics
The Hebrew University of Jerusalem
2017-2019
Doctor of Philosophy (Ph.D.)
Theoretical Physics
KTH Royal Institute of Technology
2017
Master’s Degree
Theoretical Physics
Sapienza Università di Roma
2013
Bachelor’s Degree
Physics
Sapienza Università di Roma
2010