Author
Listed:
- Yoshua Bengio
- Nicolas Chapados
Abstract
To deal with the overfitting problems that occur when there are not enough examples compared to the number of input variables in supervised learning, traditional approaches are weight decay and greedy variable selection. An alternative that has recently started to attract attention is to keep all the variables but to put more emphasis on the most useful ones. We introduce a new regularization method called "input decay"" that exerts more relative penalty on the parameters associated with the inputs that contribute less to the learned function. This method, like weight decay and variable selection, still requires to perform a kind of model selection. Successful comparative experiments with this new method were performed both on a simulated regression task and a real-world financial prediction task." Pour tenir compte des problèmes de sur-entraînement qui apparaissent quand il n'y a pas assez d'exemples comparativement au nombre de variables d'entrées durant l'apprentissage supervisé, les approches traditionnelles sont la pénalisation de la norme des paramètres (weight decay) et la sélection de variables vorace. Une alternative qui est apparue tout récemment est de garder toutes les variables, mais de mettre plus d'emphase sur celles qui sont le plus utiles. Nous introduisons une nouvelle méthode de régularisation, appelé "pénalisation sur la norme des entrées"" (input decay), qui applique une plus grande penalité relative sur les paramètres associés aux entrées qui contribuent le moins à la fonction apprise. Cette méthode, comme la pénalisation de la norme des paramètres (weight decay) et la sélection de variables, demande tout de même d'appliquer une sorte de sélection de modèle. Une série d'expériences comparatives avec cette nouvelle méthode ont été appliquées à deux taches de régression, une qui était simulée et l'autre à partir d'une vrai problème financier."
Suggested Citation
Yoshua Bengio & Nicolas Chapados, 2002.
"Input Decay: Simple and Effective Soft Variable Selection,"
CIRANO Working Papers
2002s-52, CIRANO.
Handle:
RePEc:cir:cirwor:2002s-52
Download full text from publisher
Corrections
All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:cir:cirwor:2002s-52. See general information about how to correct material in RePEc.
If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.
We have no bibliographic references for this item. You can help adding them by using this form .
If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Webmaster (email available below). General contact details of provider: https://edirc.repec.org/data/ciranca.html .
Please note that corrections may take a couple of weeks to filter through
the various RePEc services.