Luigi Malagò, PhD

ML and Opt Group | News | CV | Research Interests | Tutorials and Lectures | Publications | Downloads | Teaching Material | Links

Principal Investigator
Transylvanian Institute of Neuroscience (TINS)
Str. Ploiești 33, 400157 Cluj-Napoca, Romania
Email [my-last-name-without-any-accent]

Google Scholar
Mathematics Genealogy Project
(Last update 06 May 2021)

# Next Conferences, Meetings and News

  • Invited talk ''An Introduction to Information Geometry form the Perspective of Riemannian Optimization'' at the Virtual Meeting on Information Geometry, 2021, May 6, 2021
  • AIM Workshop: Boltzmann machines, September 17-21, 2018, American Institute of Mathematics, San Jose, CA, USA
  • Entropy 2018, May 21-24, 2018, Barcelona, Spain
  • The Ninth International Conference on Guided Self-Organisation (GSO-2018) : Information Geometry and Statistical Physics, March 26-28, 2018, Leipzig, Germany
  • GSI2017 - 3rd conference on Geometric Science of Information, November 7-9, 2017, Paris, France
  • TGSI2017 Topological and Geometrical Structure of Information - CIRM conference, August 27 - September 1, 2017, Marseille, France
  • First Italian Meeting on Probability and Mathematical Statistics, June 19 - 22, 2017, Torino, Italy
  • Dagstuhl Seminar Theory of Evolutionary Algorithms, May 7 - 12, 2017, Wadern, Germany
  • PAFT2017 Current Problems in Theoretical Physics - Information Geometry & Quantum Information, April 07 - 10, 2017, Vietri sul Mare, Italy-->
  • Cluj Innovation Days 2017 March 30-21, 2017, Cluj-Napoca, Romania
  • Workshop on the Interface of Statistics and Optimization (WISO) - SAMSI, February 08 - 10, 2017, Duke University, Durham, NC, USA
  • One-day workshop on Algebraic Statistics January 23, 2017, Department of Mathematics, Università di Genova, Italy
  • COST Action CA15140: Improving Applicability of Nature-Inspired Optimisation by Joining Theory and Practice (ImAppNIO) October 17-18, 2016, Cluj-Napoca, Romania
  • Cluj IT Days 2016 November 18-19, 2016, Cluj-Napoca, Romania
  • [03-10-2016] I just moved to the Romanian Institute of Science and Technology - RIST where I received a 4 years starting grant as Principal Investigator. POC 2014-2020 - ANCSI Competitiveness Operational Programme 2014-2020. Project name: DeepRiemann - Riemannian Optimization Methods for Deep Learning

  • Past events and older news

    # Research Interests

  • Information Geometry and its Applications to Machine Learning
  • Statistical Manifolds and Dually-flat Structures
  • First and Second-order Optimization Methods over Statistical Manifolds: Natural Gradient and alpha-Hessians
  • Stochastic Optimization and Nature-Inspired Optimisation
  • Algebraic Statistics and Exponential Families

  • Research statement

    # Short CV

  • [since 04/2021] Principal Investigator at the Transylvanian Institute of Neuroscience (TINS), Cluj-Napoca, Romania
  • [10/2016-03/2021] Project Director at the Romanian Institute of Science and Technology (RIST), Cluj-Napoca, Romania
  • [05-06/2014 and 06-07/2016] Research Assistant at Collegio Carlo Alberto, Moncalieri, Italy
  • [07/2014-04/2016] Assistant Professor at Shinshu University in Nagano, Japan
  • [10/2014-09/2015] Visiting Scientist at Inria Saclay, TAO team, Orsay Cedex, France
  • [04/2012-04/2014] Postdoc Researcher at Universita' degli Studi di Milano, Italy
  • [07/2012] Dimitris N. Chorafas Foundation Award for the best PhD thesis
  • [03/2012] PhD cum Laude in Information Technology from Politecnico di Milano, Italy
  • [07/2008] Master cum Laude in Computer Science Engineering from Politecnico di Milano / Politecnico di Torino, Italy
  • [12/2006] Diploma of Alta Scuola Politecnica
  • [10/2004] Bachelor cum Laude in Computer Science Engineering from Politecnico di Milano, Italy

  • Short biography

    # Postdocs and PhD Students

    Deepika postdoc researcher
    Hector Javier Hortua Orjuela postdoc researcher
    Riccardo Volpi postdoc researcher
    Septimia Sarbu postdoc researcher
    Sabin Roman postdoc researcher
    Alexandra Albu research assistant
    Alina Enescu research assistant and PhD student at Babeş-Bolyai University of Cluj-Napoca
    Petru Hlihor PhD student, co-supervised by Prof. Dr. Nihat Ay (MPI-MIS)
    Csongor-Huba Várady PhD student, co-supervised by Prof. Dr. Nihat Ay (MPI-MIS)

    # Grants

  • Principal Investigator of the DeepRiemann - Riemannian Optimization Methods for Deep Learning (2016-20) project. POC 2014-2020 - ANCSI Competitiveness Operational Programme 2014-2020 (~1.9M EUR)

  • # Tutorials and Lectures slides

  • Tutorial A Bridge between Optimization over Manifolds and Evolutionary Computation at PPSN 2016, Edinburgh, 17 September, 2016
  • Tutorial Information Geometry and Algebraic Statistics on a finite state space and on Gaussian models (together with Giovanni Pistone) at the Algebraic Statistics 2015, Genova, 11 June, 2015. [abstract]
  • Lecture Applications of Information Geometry to Combinatorial Optimization at the PhD Course on Information Geometry in Learning and Optimization, Copenhagen, 22-26 September, 2014. Extra material: [animation]
  • Tutorial Information Geometry in Evolutionary Computation [updated slides] (together with Tobias Glasmachers) at GECCO 2014, Vancouver, 13 July, 2014

  • # Submitted papers

  • G. Chirco, L. Malagò, and G. Pistone.
    Lagrangian and Hamiltonian Mechanics for Probabilities on the Statistical Manifold.

  • C. Várady, R. Volpi, L. Malagò, and N. Ay.
    Natural Wake-Sleep Algorithm.

  • # Recent Publications

  • R. Volpi, U. Thakur, and L. Malagò.
    Changing the Geometry of Representations: α-Embeddings for NLP Tasks.
    Entropy 2021, 23, 287, Special Issue Information Geometry III, 2021

  • R. Volpi and L. Malagò.
    Natural Alpha Embeddings.
    Information Geometry, Vol. 3, Issue 2, 2021. [arXiv:1912.02280]

  • H. J. Hortua, L. Malagò, and R. Volpi.
    Constraining the Reionization History using Bayesian Normalizing Flows.
    Machine Learning: Science and Technology, 1 035014, 2020. [arXiv:1911.08508]

  • H. J. Hortua, R. Volpi, D. Marinelli, and L. Malagò.
    Parameters Estimation for the Cosmic Microwave Background with Bayesian Neural Networks.
    Phys. Rev. D 102, 103509, 2020. [arXiv:1911.08508]

  • L. Malagò, L. Montrucchio, and G. Pistone.
    Wasserstein Riemannian Geometry of Positive Definite Matrices.
    In Information Geometry, Vol. 1, Issue 2, pp 137-179, 2018

  • Full list of papers

    # Patents

  • L. Malagò, A.-I. Albu, A. Enescu.
    Method of Detecting Anomalous Data, Machine Computing Unit, Computer Program.
    European Patent Application EP3719711, application number: 20188779.1, date of filing: 30.07.2020

  • # Downloads

  • [09-01-2014] Politecnico di Milano, DEI beamer template (based on Håvard Berland NTNU template) [zip] [pdf]

  • # Teaching Material

  • Babeș-Bolyai University, 2017-18 Datamining, Prof. Anca Andreica
  • Politecnico di Milano, 2013-14 Informatica B, Prof. Daniele Loiacono (in Italian)
  • Politecnico di Milano, 2012-13 Informatica per Ing. Civile, Prof. Salvatore Distefano (in Italian)

  • # Links

  • It’s All Latin to Me: Latin Abbreviations in Scholarly Writing by Chelsea Lee: On the differences between "cf." and "e.g.,".
  • On the differences between "besides" and "beside":
  • How to Write Goals and Objectives for Grant Proposals: using the S.M.A.R.T. method of writing your objectives.
  • Dead links checkers
  • Color scheme for your website