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

Training Linear Ranking SVMs in Linearithmic Time Using Red–Black Trees

Antti Airola, Tapio Pahikkala, Tapio Salakoski, Training Linear Ranking SVMs in Linearithmic Time Using Red–Black Trees. Pattern Recognition Letters 32(9), 1328–1336 , 2011.

Abstract:

We introduce an efficient method for training the linear ranking support vector machine. The method combines cutting plane optimization with red–black tree based approach to subgradient calculations, and has O(ms + mlog (m)) time complexity, where m is the number of training examples, and s the average number of non-zero features per example. Best previously known training algorithms achieve the same efficiency only for restricted special cases, whereas the proposed approach allows any real valued utility scores in the training data. Experiments demonstrate the superior scalability of the proposed approach, when compared to the fastest existing RankSVM implementations.

BibTeX entry:

@ARTICLE{jAiPaSa11b,
  title = {Training Linear Ranking SVMs in Linearithmic Time Using Red–Black Trees},
  author = {Airola, Antti and Pahikkala, Tapio and Salakoski, Tapio},
  journal = {Pattern Recognition Letters},
  volume = {32},
  number = {9},
  pages = {1328–1336 },
  year = {2011},
  keywords = {Bundle methods; Cutting plane method; Learning to rank; Ranking support vector machine; Red–black tree},
}

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

Publication Forum rating of this publication: level 2

Edit publication