Séminaires RALI-OLST

A User's Guide to AdaBoost

Balazs Kegl


Le mercredi 31 mars 2004 à 11 h 30

Salle 3195, Pavillon André-Aisenstadt

AdaBoost and Support Vector Machines (SVM) are two classification algorithms that have been revolutionizing the technology of pattern recognition since the late nineties. Both algorithms require considerably less "engineering" and parameter tuning than Multilayer Neural Networks, their predecessor that ruled applied pattern recognition in the late eighties and early nineties. The main objective of this talk is to describe AdaBoost and some of its variants from a practitioner's point of view. First, I will outline the main intuitions behind the algorithm, provide the algorithmic details, and give a short demonstration on a two-dimensional dataset to illustrate the main ideas. Then I will show some of the most important extensions and variants of the basic algorithm. Finally, I will discuss the design of the expert pool through some applications in the areas of image and language processing.

Pour recevoir les annonces hebdomadaires par courriel, envoyez un message à l'adresse majordomo@iro.umontreal.ca. Il suffit d'envoyer un message ne contenant que la ligne 'subscribe ralli' (sans les apostrophes, avec un double 'l' dans 'ralli').

Liste de tous les séminaires pour l'année :

1991 1992 1993 1994 1995 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022