| HAL : hal-00643001, version 1 |
| Voir la fiche détaillée | BibTeX,EndNote,... |
|
|
| Yahoo! Learning to Rank Challenge, Haifa : Israël (2010) |
|
|
|
|
| Ranking by calibrated AdaBoost |
|
|
| Róbert Busa-Fekete1Balázs Kégl1, 2, 3Tamas Elteto3György Szarvas4 |
|
|
| (15/06/2011) |
|
|
| This paper describes the ideas and methodologies that we used in the Yahoo learning-to- rank challenge1. Our technique is essentially pointwise with a listwise touch at the last combination step. The main ingredients of our approach are 1) preprocessing (querywise normalization) 2) multi-class AdaBoost.MH 3) regression calibration, and 4) an expo- nentially weighted forecaster for model combination. In post-challenge analysis we found that preprocessing and training AdaBoost with a wide variety of hyperparameters im- proved individual models significantly, the final listwise ensemble step was crucial, whereas calibration helped only in creating diversity. |
|
|
|
|
|
|
|
|
|
|
| 1 : | LAL - Laboratoire de l'Accélérateur Linéaire |
| 2 : | LRI - Laboratoire de Recherche en Informatique |
| 3 : | INRIA Saclay - Ile de France - TAO |
| 4 : | Ubiquitous Knowledge Processing (UKP) Lab |
|
|
|
|
|
|
|
|
| Domaine | : | Informatique/Apprentissage |
|
|
| Liste des fichiers attachés à ce document : | |||||
|
|
|
| hal-00643001, version 1 | |
| http://hal.inria.fr/hal-00643001 | |
| oai:hal.inria.fr:hal-00643001 | |
| Contributeur : Balázs Kégl | |
| Soumis le : Dimanche 20 Novembre 2011, 22:56:57 | |
| Dernière modification le : Mardi 30 Octobre 2012, 17:13:36 | |