Affiliation at the University of Liège:
Machine learning, in
particular tree based models, graphical models, and reinforcement
learning, with a focus on complex and large
analysis, optimization and control.
Electric power systems reliability management and optimal decision making under uncertainties (system operation, asset management, grid development).
Bioinformatics and computational systems biology, in particular biological systems and structures modeling, exploitation of proteomics, genomics, clinical and biomedical imaging data.
Eléments du calcul des probabilités (ULg)
Eléments de statistique (ULg)
Information and coding theory: (ULg)
Introduction to machine learning: (ULg)
Advanced machine learning: (ULg)
Intelligent Robotics: (ULg)
ORBI: University of Liège Repository of Publications.
DTU 2018: Machine Learning for Probabilistic Power Systems Reliability Management Seminar at the DTU (Danish Technical University), Lyngby, Nov. 23, 2018. (7.3 Mbytes)
LIST 2016: Big data, machine learning, and optimization, for power systems reliability Workshop at the LIST (Luxembourg Institute of Technology), Belvaux, Nov. 9, 2016 (10.2 Mbytes)
NRC 2015: How to combine observational data sources with first principles of physics to build stable and transportable models for power system design and control? Plenary presentation at "Analytic Research Foundations for the Next-Generation Grid". A workshop of the National Research Council of the National Academies, Irvine Feb. 11-12, 2015 (10.2 Mbytes)
PSCC 2014: Adavanced optimization for power systems, Plenary survey presentation at 18th PSCC - Wroclaw, August 20, 2014 (4.6 Mbytes)
LRI-Orsay-2011 : Regression tree ensembles in the perspective of kernel-based methods, Laboratoire de Recherche en Informatique - Paris 11 - Orsay, Avril 23, 2011 (1.9 Mbytes)
MPI-Tuebingen-2009 : Regression tree ensembles in the perspective of kernel-based methods, Max Planck Insitute for Biological Cybernetics - Tuebingen, October 30, 2009 (1.9 Mbytes)
IAP V Study day 05 : Decision and regression tree ensemble methods and their applications in automatic learning, IAP V Study day, Colonster, May 19, 2005 (2563581 bytes)
ORBEL05 : Decision and regression tree ensemble methods and their applications in automatic learning, ORBEL Symposium on data mining, Brussels, March 16, 2005 (2596449 bytes)
EC-ICT05 : Closure of session 3, The future of ICT for power systems: emerging security challenges, European Commission, Brussels, February 3-4, 2005 (441856 bytes).
IREP2004 : Whither dynamic congestion management?, IREP Workshop, Contrina d'Ampezzo, August 2004. (98816 bytes)
CBRN2001 : Recent developments in tree induction for KDD. «Towards soft tree induction», Brasilian conference on Neural Networks, Rio, April 2001. (1592832 bytes)
PICA99 : Automatic learning and data mining applied to security assessment, PICA99 panel session, Santa Clara (Ca), May 1999, slides powerpoint gzipped (765770 bytes).
IBM-ARC 99 and IFSA97: Discretization of continuous attributes for supervised learning. Variance evaluation and variance reduction, May 1999, slides pdf (135148 bytes).
IBM-ARC 99 and IPMU92 : A global tree quality measure and its use for pruning, May 1999, slides pdf (123098 bytes).
LESCOPE98 : Visualizing Dynamic Power System Scenarios for Data Mining, LESCOPE98, Halifax (NS), June 1998, slides (351583 bytes).
IEEEWM98 : Artificial Intelligence Methods for Voltage Stability Assessment, IEEE PES Winter Meeting, Tampa (Fl), February 1998, slides of presentation to the Power System Stability Subcommittee (452086 bytes).
KDDLyon97 : Data mining and KDD Winter School, University of Lyon, Lyon (Fr), December 1997, slides of presentation (813919 bytes).
CPSPP97 : Tutorial on Intelligent Systems and their Power System applications, IFAC-Cigré Symp. on Control of Power Systems and Power Plants, Beijing (PRC), August 1997, course notes (330583 bytes).
PICA97 : Tutorial on Automatic Learning Methods. Application to Dynamic Security Assessment, IEEE Power Industry Computer Applications Conference, Columbus (Oh), May 1997, course notes (530313 bytes).