Scientific Program and Schedule


September 5, 2018

17.00-18.30     Registration

18.00-18.15     Opening


18.15-19.40     Talk Session 1
Robustness and regularization
Chair: Christian Hennig

  • Andrea Cerioli (Italy), Alessio Farcomeni, Marco Riani
    Wild adaptive trimming for robust estimation and cluster analysis
  • Roberto Di Mari, Roberto Rocci (Italy), Stefano Antonio Gattone
    Constrained maximum likelihood estimation of clusterwise linear regression models with unknown number of components
  • Francesca Greselin (Italy), Luis Angel García-Escudero, Agustin Mayo-Iscar
    Exploring robust fuzzy clustering on multivariate skew data


19.40-21.15     Welcome Cocktail

September 6, 2018

08.00-08.30     Registration

08.30-09.25     Keynote Lecture 1
Dimitris Karlis (Athens University of Economics and Business, Greece)
Model Based Clustering through copulas for high dimensional data
Chair: Maurizio Vichi



09.25-10.20     Keynote Lecture 2
Peter Bühlmann (ETH Zürich, CH)
Heterogeneity in large-scale data: invariance, causality and novel robustness
Chair: Hajo Holzmann



10.20-10.50     Coffee Break


10.50-12.35     Talk Session 2
Supervised and unsupervised classification
Chair: Luis Angel García-Escudero

  • Michael Fop (University College Dublin, Ireland), Pierre-Alexandre Mattei, Thomas Brendan Murphy, Charles Bouveyron
    Unobserved classes and extra variables in high-dimensional discriminant analysis
  • Daniel R. Jeske (University of California Riverside, USA)
    Maximizing the usefulness of statistical classifiers for two populations
  • Christian Hennig (University College London, UK)
    A multivariate characterisation of some popular cluster analysis methods
  • Carlo Cavicchia, Maurizio Vichi (Sapienza University of Rome, Italy), Giorgia Zaccaria
    Model-based double hierarchical parsimonious clustering


12.35-13.00     Lightning Talk Session 1
Chair: Keefe Murphy
  • Christophe Biernacki, Alexandre Lourme (University of Bordeaux, France)
    Data Units as (Co-)Clustering Model Enlargement
  • Andrea Cappozzo (University of Milano-Bicocca, Italy), Francesca Greselin, Thomas Brendan Murphy
    Robust adaptive eigenvalue decomposition discriminant analysis: supervised learning in presence of outliers, label noise and unobserved classes
  • Andrea Cerioli, Domenico Perrotta (European Commission, Joint Research Centre), Luis Angel García-Escudero, Agustin Mayo-Iscar, Francesca Torti
    A thinned trimmed CEM algorithm for robust clustering around regression lines
  • Przemysław Jaśko (Crakow University of Economics, Poland), Daniel Kosiorowski and Ewa Szlachtowska
    Critical discussions ofselected robust clustering procedures and their applications in Economics
  • Volodymyr Melnykov, Yana Melnykov (The University of Alabama, USA), Xuwen Zhu
    Robust clustering in the presence of skewed data groups
  • Volodymyr Melnykov, Shuchismita Sarkar, Rong Zheng (Western Illinois University, USA)
    Gaussian mixture modeling and model-based clustering under measurement inconsistency
  • Marco Riani, Francesca Torti (European Commission, Joint Research Centre), Domenico Perrotta
    Simulating mixtures of non-normal multivariate data with fixed cluster overlap in FSDA.


13.00-14.00     Lunch


14.00-15.00     Talk Session 3
Clustering of discrete data
Chair: Roberto Rocci
  • Julien Jacques (Université de Lyon, France), Margot Selosse, Christophe Biernacki
    Analyzing large matrices of ordinal data
  • Ann Ziwen, H. Al-Wahsh, Abdulkadir Hussein (University of Windsor, Canada)
    Zero-inflated count models with application to hospital performance measures

15.00-15.55     Keynote Lecture 3
Eustasio del Barrio (University of Valladolid, Spain)
Unifying Robust clustering aggregation based on optimal transportation
Chair: Elisabeth Gassiat



15.55-16.25     Coffee Break


16.25-17.50     Talk Session 4
Model-based clustering of complex data
Chair: Claire Gormley
  • Silvia D'Angelo (Sapienza University of Rome, Italy), Michael Fop, Marco Alfò
    Clustering for multidimensional networks via infinite mixture models
  • Antonello Maruotti (Libera Università  Maria SS. Assunta, Italy), Antonio Punzo, Jan Bulla
    On modelling multivariate high-dimensional time series: a factorial hidden Markov model
  • Vincent Vandewalle (Université Lille 2, France)
    Model based clustering of spatial functional data


20.30     Workshop Dinner (bus leaving at 18.45)

September 7, 2018

8.45-10.10     Talk Session 5
Recent developments in clustering of matrix data
Chair: Volodymyr Melnikov

  • Volodymyr Melnykov, Xuwen Zhu (University of Louisville, USA)
    Matrix transformation mixture modeling
  • Michael P.B. Gallaugher, Paul D. McNicholas (McMaster University, Canada)
    Mixtures of matrix variate bilinear Factor Analyzers
  • Shuchismita Sarkar, Volodymyr Melnykov (The University of Alabama, USA), Xuwen Zhu
    Model-based clustering of tensor data


10.10-10.40     Lightning Talk Session 2
Chair: Salvatore Daniele Tomarchio
  • Stefano Barberis (University of Milano-Bicocca, Italy), Salvatore Ingrassia, Giorgio Vittadini
    Generalized additive cluster weighted models
  • Alessandro Casa (Università di Padova, Italy), Luca Scrucca, Giovanna Menardi
    Averaging via stacking in model-based clustering
  • Nam-Hwui Kim (University of Waterloo, Canada), Ryan Browne
    Subspace clustering for the finite fixture of generalized hyperbolic distributions
  • Gertraud Malsiner-Walli (WU Vienna University of Business and Economics, Austria), Sylvia Frühwirth-Schnatter, Bettina Grün
    Learning the number of components and data clusters in Bayesian finite mixture models
  • Gildas Mazo (Université Paris-Saclay, France), Yaroslav Averyanov
    Constraining kernel estimators in semiparametric copula-based mixture models
  • Keefe Murphy (University College Dublin, Ireland), Thomas Brendan Murphy
    Gaussian parsimonious clustering models with covariates
  • Marta Nai Ruscone (LIUC Università Cattaneo, Italy), Thomas Brendan Murphy
    Model-based clustering with R-vine copulas


10.40-11.10     Coffee Break


11.10-12.05     Keynote Lecture 4
Cinzia Viroli (University of Bologna, Italy)
Deep Gaussian Mixture Models
Chair: Christophe Biernacki


12.05-13.00     Keynote Lecture 5
Nello Cristianini (University of Bristol, UK)
Things to keep in mind when designing Intelligent Algorithms
Chair: Angela Montanari


13.00-14.00     Lunch


14.00-15.25     Talk Session 6
Issues in hidden Markov models
Chair: Antonello Maruotti
  • Yohann De Castro, Sylvain Le Corff, Elisabeth Gassiat (Université Paris-Sud, France)
    Consistent estimation of the filtering and smoothing probabilities in non parametric hidden Markov models
  • Hajo Holzmann (Philipps-Universität Marburg, Germany), Anna Leister, Grigory Alexandrovich, Ann-Kristin Becker
    Time-dependent nonparametric latent variable modeling
  • Alessio Farcomeni (Sapienza University of Rome, Italy), Gordon Anderson, Maria Grazia Pittau, Roberto Zelli
    Time-specific clustering via rectangular latent Markov models, with an analysis of the well being of nations


15.25-15.50     Lightning Talk Session 3
Chair: Przemysław Jaśko
  • Laura Anderlucci, Roberta Falcone (University of Bologna, Italy), Angela Montanari
    Supervised classification with matrix sketching
  • Leopoldo Catania (Aarhus University, Denmark), Roberto Di Mari, Paolo Santucci de Magistris
    The analysis of high frequency financial price changes
  • Majed El Helou, Rawan Chanouha (AUB, Beirut, Lebanon), Hazem Hajj
    Multi-resolution bagging for ensemble Classification
  • Mario Fordellone (Sapienza University of Rome, Italy), Maurizio Vichi
    Improving clustering assessment through supervised classification modeling
  • Shuchismita Sarkar (The University of Alabama, USA), Xuwen Zhu, Volodymyr Melnykov, Salvatore Ingrassia
    Parsimonious models in matrix data mixture modeling
  • Zdenek Sulc (University of Economics, Prague, Czech Republic), Beibei Juan, Mark de Rooij
    The Delta Machine: binary data classification
  • Salvatore Daniele Tomarchio (Università di Catania, Italy), Antonio Punzo
    Zero-and-one inflated mixtures for loss given default


15.50-16.50     Poster Session with Coffee Break


16.50-18.15     Talk Session 7
Developments in modeling high-dimensional data
Chair: Paul D. McNicholas
  • Laura Anderlucci (University of Bologna, Italy), Francesca Fortunato, Angela Montanari
    High-dimensional clustering with Random Projections
  • Pietro Coretto (Università di Salerno, Italy)
    Robust patients sub-typing with noisy high-dimensional gene expression data
  • Keefe Murphy, Isobel Claire Gormley (University College Dublin, Ireland), Cinzia Viroli
    Infinite mixtures of infinite factor analysers


18.15-18.20     Closing


18.20-19.45     Wine Session
South-East taste of wine: the Greek soul of Sicily

Chair: Mariella Ferrara