Exploiting Ontology Structures and Unlabeled Data for Learning

Nina Balcan, Avrim Blum, Yishay Mansour
Proceedings of the 30th International Conference on Machine Learning, PMLR 28(3):1112-1120, 2013.

Abstract

We present and analyze a theoretical model designed to understand and explain the effectiveness of ontologies for learning multiple related tasks from primarily unlabeled data. We present both information-theoretic results as well as efficient algorithms. We show in this model that an ontology, which specifies the relationships between multiple outputs, in some cases is sufficient to completely learn a classification using a large unlabeled data source.

Cite this Paper

BibTeX
@InProceedings{pmlr-v28-balcan13a, title = {Exploiting Ontology Structures and Unlabeled Data for Learning}, author = {Balcan, Nina and Blum, Avrim and Mansour, Yishay}, booktitle = {Proceedings of the 30th International Conference on Machine Learning}, pages = {1112--1120}, year = {2013}, editor = {Dasgupta, Sanjoy and McAllester, David}, volume = {28}, number = {3}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/balcan13a.pdf}, url = {https://proceedings.mlr.press/v28/balcan13a.html}, abstract = {We present and analyze a theoretical model designed to understand and explain the effectiveness of ontologies for learning multiple related tasks from primarily unlabeled data. We present both information-theoretic results as well as efficient algorithms. We show in this model that an ontology, which specifies the relationships between multiple outputs, in some cases is sufficient to completely learn a classification using a large unlabeled data source.} }
Endnote
%0 Conference Paper %T Exploiting Ontology Structures and Unlabeled Data for Learning %A Nina Balcan %A Avrim Blum %A Yishay Mansour %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-balcan13a %I PMLR %P 1112--1120 %U https://proceedings.mlr.press/v28/balcan13a.html %V 28 %N 3 %X We present and analyze a theoretical model designed to understand and explain the effectiveness of ontologies for learning multiple related tasks from primarily unlabeled data. We present both information-theoretic results as well as efficient algorithms. We show in this model that an ontology, which specifies the relationships between multiple outputs, in some cases is sufficient to completely learn a classification using a large unlabeled data source.
RIS
TY - CPAPER TI - Exploiting Ontology Structures and Unlabeled Data for Learning AU - Nina Balcan AU - Avrim Blum AU - Yishay Mansour BT - Proceedings of the 30th International Conference on Machine Learning DA - 2013/05/26 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-balcan13a PB - PMLR DP - Proceedings of Machine Learning Research VL - 28 IS - 3 SP - 1112 EP - 1120 L1 - http://proceedings.mlr.press/v28/balcan13a.pdf UR - https://proceedings.mlr.press/v28/balcan13a.html AB - We present and analyze a theoretical model designed to understand and explain the effectiveness of ontologies for learning multiple related tasks from primarily unlabeled data. We present both information-theoretic results as well as efficient algorithms. We show in this model that an ontology, which specifies the relationships between multiple outputs, in some cases is sufficient to completely learn a classification using a large unlabeled data source. ER -
APA
Balcan, N., Blum, A. & Mansour, Y.. (2013). Exploiting Ontology Structures and Unlabeled Data for Learning. Proceedings of the 30th International Conference on Machine Learning, in Proceedings of Machine Learning Research 28(3):1112-1120 Available from https://proceedings.mlr.press/v28/balcan13a.html.

Related Material