Inference with the Universum

Weston, J; Collobert, R; Sinz, F; Bottou, L; Vapnik, V

doi:10.1145/1143844.1143971

アイテム詳細

登録内容を編集ファイル形式で保存

一時保存へ追加

タグ情報を表示リリース履歴を表示詳細要約

公開

会議論文

Inference with the Universum

MPS-Authors

/persons/resource/persons84226

Sinz, F
Max Planck Institute for Biological Cybernetics, Max Planck Society;
Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society;

External Resource

https://dl.acm.org/citation.cfm?doid=1143844.1143971
(出版社版)

Fulltext (restricted access)

There are currently no full texts shared for your IP range.

フルテキスト (公開)

ICML-2006-Weston.pdf
(全文テキスト（全般）), 393KB

付随資料 (公開)

There is no public supplementary material available

引用

Weston, J., Collobert, R., Sinz, F., Bottou, L., & Vapnik, V. (2006). Inference with the Universum. In W., Cohen, & A., Moore (Eds.), ICML '06: Proceedings of the 23rd International Conference on Machine Learning (pp. 1009-1016). New York, NY, USA: ACM Press.

引用: https://hdl.handle.net/11858/00-001M-0000-0013-D139-2

要旨

WIn this paper we study a new framework introduced by Vapnik (1998) and Vapnik (2006) that is an alternative capacity concept to the large margin approach. In the particular case of binary classification, we are given a set of labeled examples, and a collection of "non-examples" that do not belong to either class of interest. This collection, called the Universum, allows one to encode prior knowledge by representing meaningful concepts in the same domain as the problem at hand. We describe an algorithm to leverage the Universum by maximizing the number of observed contradictions, and show experimentally that this approach delivers accuracy improvements over using labeled data alone.