BEGIN:VCALENDAR VERSION:2.0 PRODID:-//wp-events-plugin.com//7.2.3.1//EN TZID:Europe/Paris X-WR-TIMEZONE:Europe/Paris BEGIN:VEVENT UID:7912@i2m.univ-amu.fr DTSTART;TZID=Europe/Paris:20160201T000000 DTEND;TZID=Europe/Paris:20160205T000000 DTSTAMP:20241221T200358Z URL:https://www.i2m.univ-amu.fr/evenements/statistical-learning-thematic-m onth-2016/ SUMMARY:Conference (CIRM\, Luminy\, Marseille): Statistical learning (Thema tic Month 2016) DESCRIPTION:Conference: \n\n\n\n\n\n\n\n\n\n\n\n\n Schedule \n\n\n\n\n\n Li st of participants \n\n\n\n\n\n Sponsors \n\n\n\n\n\n Abstracts \n\n\n\n\n \n Slides \n\n\n\n\n\n Videos \n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nTHEMATIC MON TH 2016\nWeek 1: Statistical learning\nFebruary 1 - 5\, 2016\n\n\n\n\n\n\n \nThis week will be devoted to statistical learning from both the theoreti cal and applied perspectives. Statistical learning theory has been develop ed in the 70's and brought a great revival in statistics. On the one hand\ , the development of computer science and tools allowed massive data colle ction\, and implementation of powerful algorithms which are often memory a nd computation time consuming. On the other hand\, classical asymptotic th eory used to prove the efficiency of estimation methods in modeling and pr ediction was limited by dimensionality problems.\nThe approaches developed in statistical learning helped to face some new challenges such as the cu rse of dimensionality\, small sample size\, and now massive  datasets. Da ta mining\, feature selection\, and more recently Big data appeared as spe cialized approaches for massive datasets modeling and analysis.\nIn additi on to the theoretical developments (non asymptotic theory)\, many algorith ms emerged in statistics and computer science to meet these new needs whic h arised in many areas such as bioinformatics\, social sciences\, medicine or telecommunications.\nThis week aims to bring together specialists from statistical learning working on advanced techniques and coming from diffe rent fields\, mainly statistics\, but also computer science\, social scien ce and bioinformatics.\nThe main topics of interest include:\n\n Supervis ed learning\n Unsupervised learning: clustering and density estimation\n Model selection\n Big data\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nScientific Commi ttee\n\nGérard Biau (Université Pierre-et-Marie-Curie)\nPascal Massart ( Université Paris-Sud)\nVincent Rivoirard (Université Paris-Dauphine)\n\n Organizing Committee\n\nBadih Ghattas (Aix-Marseille Université)\nLiva Ra laivola (Aix-Marseille Université)\n\nSpeakers\n\n Francis Bach (ENS Par is\, France)\n\nLarge-scale machine learning and convex optimization\n\n Philippe Besse (INSA Toulouse)\n\nApprentissage et données massives\n\n Gilles Blanchard (Potsdam University)\n\nIs adaptive early stopping possib le in statistical inverse problems?\n\n Sébastien Bubeck (Princeton Univ ersity &\; Microsoft Seattle)\n\nEntropy\, geometry\, and a CLT for Wis hart matrices\n\n Peter Buhlmann (ETH Zurich)\n\nThe power of heterogeneo us large-scale data for high-dimensional causal inference\n\n Stéphane C anu (INSA Rouen)\n\nMixed integer programming for sparse and non convex ma chine learning\n\n Stéphane Chrétien (National Physical Laboratory\, Te ddington\, UK)\n\nA Lagrangian viewpoint on Robust PCA\n\n Emilie Devijve r (Orsay-Belgique)\n\nBlock-diagonal covariance selection for high-dimensi onal Gaussian\ngraphical models\n\n Stéphane Gaïffas (Ecole polytechniq ue)\n\nStatistical learning with Hawkes processes and new matrix concentra tion inequalities\n\n Pierre Geurts (Université de Liège)\n\nRandom for ests variable importances: towards a better understanding and large-scale feature selection\n\n Claire Lacour (Université Paris-Sud)\n\nAbout the Goldenshluger-Lepski methodology for bandwidth selection\n\n Matthieu Ler asle (Université Nice-Sophia-Antipolis)\n\nSub-Gaussian mean estimators\n \n Clément Levrard (Université Paris Diderot)\n\nReconstruction simplic iale de variétés via l'estimation des plans tangents\n\n  Sébastien L oustau (Université d'Angers)\n\nQuantization\, Learning and Games with OP AC\n\n Stéphane Mallat (Ecole polytechnique)\n\nUnderstanding (or not) D eep Neural Networks\n\n André Mas (Université de Montpellier)\n\nEigenv alue-free risk bounds for PCA projectors\n\n Patricia Reynaud-Bouret (Uni versité Nice-Sophia-Antipolis)\n\nEstimation  of  local  independence   graphs  via  Hawkes  processes  and link with the  functional neur onal connectivity\n\n Eric Sibony (Telecom Paris Tech)\n\nA novel multi r esolution framework for the statistical analysis of ranking data\n\n Gill es Stoltz (CNRS\, HEC Paris\, France)\n\nRobust sequential learning with a pplications to the forecasting of electricity consumption and of exchange rates\n\n Alexandre Tsybakov (CREST-ENSAE)\n\nOracle inequalities for net work models and sparse graphon estimation\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\ n\n\n \n\n\n\n\nTRUSTEES \n\n\n\n\n\n\n\n\n\n\n\n \n\n\n\n\n\n\n\n\n \ n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\ n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n \n\n\n\n\n\n\n\n\n\n\n\n\n\n CATEGORIES:Colloque,Mois thématique END:VEVENT BEGIN:VTIMEZONE TZID:Europe/Paris X-LIC-LOCATION:Europe/Paris BEGIN:STANDARD DTSTART:20151025T020000 TZOFFSETFROM:+0200 TZOFFSETTO:+0100 TZNAME:CET END:STANDARD END:VTIMEZONE END:VCALENDAR