Machine Learning and Optimization Group at RIST

Romanian Institute of Science and Technology - RIST
Str. Virgil Fulicea nr. 17, 400022 Cluj-Napoca, Romania

# Open Positions

Postdoc positions on Machine Learning, Deep Learning, Optimization, and Information Geometry

  • 1 postdoc position on Deep Learning and Machine Learning, starting fall 2018
  • 1 postdoc position on Machine Learning, Optimization, Deep Learning and Information Geometry, starting fall 2018
  • 1 open position for research assistant / PhD student on Deep Learning and Machine Learning, starting fall 2018
  • The lab has open positions for Summer Internships for Universitatea Babeș-Bolyai (UBB) and Universitatea Tehnica din Cluj-Napoca (TU) students, starting from July 2018

  • # News

  • [19-05-2017] Prof. Giovanni Pistone, from Collegio Carlo Alberto, Italy, will visit our group from 8 to 14 May 2017.
  • [18-05-2017] PhD Deepika Kumari, from Guru Gobind Singh Indraprastha University, India, will visit our group from 3 to 7 July 2017. She will have a public seminar on the 04.07.2017 entitled "Geometry of Biharmonic Submanifolds".
  • [16-05-2017] Alexandra and Titus will participate to the Data Science Summer School 2017, at École Polytechnique, from 28 August to 1 September 2017. Alexandra will present a poster on variational autocoders and Titus on geometric methods to train neural networks.

  • # Research Interests

  • Deep Learning
  • Machine Learning
  • Information Geometry
  • Riemannian and Stochastic Optimization
  • Reinforcement Learning

  • # Members

    Luigi Malagò lastname-without-any-accent[AT]rist[DOT]ro group leader
    Deepika name[AT]rist[DOT]ro postdoc researcher
    Dimitri Marinelli lastname[AT]rist[DOT]ro postdoc researcher
    Riccardo Volpi lastname[AT]rist[DOT]ro postdoc researcher
    Septimia Sârbu lastname[AT]rist[DOT]ro postdoc researcher
    Sabin Roman lastname[AT]rist[DOT]ro postdoc researcher
    Petru Hlihor lastname[AT]rist[DOT]ro phd student, co-supervised by Prof. Dr. Nihat Ay (MPI-MIS)
    Alexandra Pește lastname-without-any-accent[AT]rist[DOT]ro phd student, co-supervised by Prof. Dr. Nihat Ay (MPI-MIS)
    Larisa Calo firstname[DOT]lastname[AT]rist[DOT]ro project assistant

    # Internships Students

    Alexandra Albu July-August 2017; from July 2018 master internship student from Babeș-Bolyai University, Cluj-Napoca
    Marco Ciccone May-July 2018 PhD internship student from Politecnico di Milano, Italy
    Roxana Jurca spring 2017 bachelor internship student from Technical University Cluj-Napoca

    # Former Members

    Titus Nicolae research assistant (February-July 2017)
    Raluca Drozan project assistant (December 2016 - December 2017)

    # Publications

    Papers in International Journals

  • L. Malagò, L. Montrucchio, and G. Pistone.
    Wasserstein Riemannian Geometry of Positive Definite Matrices.
    Accepted in Information Geometry, 2018

  • Papers in Workshops

  • S. Sârbu, R. Volpi, A. Pește, and L. Malagò.
    Learning in Variational Autoencoders with Kullback-Leibler and Renyi Integral Bounds.
    In ICML 2018 Workshop on Theoretical Foundations and Applications of Deep Generative Models, Stockholm, Sweden, 14-15 July 2018.

  • A. Pește, L. Malagò, and S. Sârbu.
    An Explanatory Analysis of the Geometry of Latent Variables Learned by Variational Auto-Encoders.
    In NIPS 2017 Workshop on Bayesian Deep Learning, Long Beach, US, 9 December 201.

  • A. Pește and L. Malagò.
    Towards the Use of Gaussian Graphical Models in Variational Autoencoders.
    In ICML 2017 Workshop on Implicit Models, Sydney, Australia, 10 August 2017.

  • L. Malagò and D. Marinelli.
    Synthetic Generation of Local Minima and Saddle Points for Neural Networks.
    In ICML 2017 Workskop on Principled Approaches to Deep Learning, Sydney, Australia, 10 August 2017.

  • # Grants

    The Machine Learning and Optimization Group is supported by the POC 2014-2020 grant - ANCSI Competitiveness Operational Programme 2014-2020. Project name: DeepRiemann - Riemannian Optimization Methods for Deep Learning (2016-20) (~1.9M EUR) Principal Investigator Dr. Luigi Malagò

    The DeepRiemann project aims at the design and analysis of novel training algorithms for Neural Networks in Deep Learning, by applying notions of Riemannian optimization and differential geometry. The task of the training a Neural Network is studied by employing tools from Optimization over Manifolds and Information Geometry, by casting the learning process to an optimization problem defined over a statistical manifold, i.e., a set of probability distributions. The project is highly interdisciplinary, with competences spanning from Machine Learning to Optimization, Deep Learning, Statistics, and Differential Geometry. The objectives of the project are multiple and include both theoretical and applied research, together with industrial activities oriented to transfer knowledge, from the institute to a startup or spin-off of the research group.

    # How to reach us

    The office is located in the city center of Cluj, only 9km away from the Cluj International Airport (Aeroportul International Avram Iancu Cluj). The simplest way to reach our office from the airport is to take a Taxi (around 25 lei). Alternatively, you can take the buses n. 5 or 8 in front of the airport, both stop at less than 10 minutes by walk from the office (ticket 2 lei). More information about the location of the bus stops can be found using Google Maps.

    Back to the homepage

    (Last update 20 August 2018)