• TUCS Publication Series
• How to Publish in TUCS Series
• Publication Search
• Publication Graph
• Publication Input
• Publication Input Guide
• JuFo Browser

Learning Preferences with Co-Regularized Least-Squares

Evgeni Tsivtsivadze, Fabian Gieseke, Tapio Pahikkala, Jorma Boberg, Tapio Salakoski, Learning Preferences with Co-Regularized Least-Squares. In: Proceedings of the ECML/PKDD Workshop on Preference Learning (PL-08), 2008.

Abstract:

Situations when only a limited amount of labeled data and a large amount of unlabeled data is available to the learning algorithm are typical for many real-world problems. In this paper, we propose a semi-supervised preference learning algorithm that is based on the multi-view approach. Multi-view learning algorithms operate by constructing a predictor for each view and by choosing such prediction hypotheses that minimize the
disagreement among all of the predictors on the unlabeled data. Our
algorithm, that we call Sparse Co-RankRLS, stems from the single-view preference learning algorithm RankRLS. It minimizes a least-squares approximation of the ranking error and is formulated within the co-regularization framework. The experiments demonstrate a
significantly better performance of Sparse Co-RankRLS compared to the
standard RankRLS algorithm. Moreover, our semi-supervised preference
learning algorithm has a linear complexity in the number of unlabeled
data items, making it applicable to large datasets.

Files:

Full publication in PDF-format

BibTeX entry:

@INPROCEEDINGS{inpTsGiPaBoSa08a,
  title = {Learning Preferences with Co-Regularized Least-Squares},
  booktitle = {Proceedings of the ECML/PKDD Workshop on Preference Learning (PL-08)},
  author = {Tsivtsivadze, Evgeni and Gieseke, Fabian and Pahikkala, Tapio and Boberg, Jorma and Salakoski, Tapio},
  year = {2008},
  keywords = {preference learning, co-regularization},
}

Belongs to TUCS Research Unit(s): Turku BioNLP Group

Edit publication