ESSEC Working Group on Risk

L'Institut des Actuaires s'associe au Working Group on Risk de l'ESSEC-CREAR pour sa prochaine conférence sur le thème : " Random Forests for Big Data " par le Professeur Jean-Michel POGGI de l’Université. Paris Descartes et LMO, Université. Paris Sud

La conférence aura lieu le vendredi 9 novembre à 12h30 à l’ESSEC campus La Défense (CNIT) - Salle 202.

Attention, nombre de places limité. Participer à cette conférence vous permettra d'ajouter 6 points à votre score de Perfectionnement Professionnel Continu (PPC).

Short bio of the speaker: 

Jean-Michel Poggi is Professor of Statistics at Paris-Descartes University and Lab. Maths Orsay University. His research interests are in time series, wavelets, tree-based and resampling methods, applied statistics. Research activities combine theoretical and practical contributions together with industrial applications (mainly environment and energy) and software development.

He is Elected Member of the ISI. He is Associate Editor of three journals: Journal of Statistical Software, CSBIGS (Case Studies in Business, Industry and Government Statistics) and Journal de la SFdS.
From 2011 to 2013, he was President of the French Statistical Society (SFdS). He is Vice-President of ECAS, member of the Board of Directors of the ERS of IASC and Council Member of the ISI.

Abstract: Big Data is a major challenge of statistical science and has numerous algorithmic and theoretical consequences. Big Data always involves massive data and often includes online data and data heterogeneity. Recently statistical methods have been adapted to process Big Data, like linear regression models, clustering methods and bootstrapping schemes. Based on decision trees combined with aggregation and bootstrap ideas, random forests (RF) are a powerful nonparametric statistical method allowing to consider in a single and versatile framework regression problems, as well as classification ones. Focusing on classification problems, this talk proposes a review of proposals that deal with scaling random forests to Big Data problems. These proposals rely on parallel environments or on online adaptations of RF. We also describe how the out-of-bag error is addressed in these methods. Then, we formulate various remarks for RF in the Big Data context. Finally, we experiment five variants on two massive datasets, a simulated one and a real-world dataset. These numerical experiments lead to highlight the relative performance of the different variants, as well as some of their limitations.This talk is related to the paper:R. Genuer, J-M. Poggi, C. Tuleau-Malot, N. Villa-Vialaneix, Random Forests for Big Data, Big Data Research, 9, 2017, 28-46.

A l'initiative du Centre de Recherche sur le Risque de l'ESSEC – CREAR, avec le soutien de l'Institut des Actuaires, du LabEx MME-DII, du Ceressec et du bureau Banque-Finance-Assurance de la Société Française de Statistique, le WG Risk organise environ deux fois par mois des conférences sur le thème des risques. Réunissant professionnels, universitaires et étudiants, et coordonnées par Marie Kratz, Professeure à l'ESSEC, les conférences données par des professionnels ou des académiques sur des sujets d'actualité peuvent être en français ou en anglais. Retrouvez toutes les informations pratiques et les précédentes conférences sur le site de CREAR-ESSEC : http://crear.essec.edu/working-group-on-risk