BibSonomy :: bibtex  ::

tag user group author concept BibTeX key search:all search:ldietz
A blue social bookmark and publication sharing system.
tags · relations · groups · popular
help · blog · about
login · register
ldietz's BibTeX entry:  

Transductive Learning via Spectral Graph Partitioning

Proceedings of the Twentieth International Conference on Machine Learning (ICML-2003), 2003.
Authors: Thorsten Joachims
URL: http://www.cs.cornell.edu/People/tj/publications/joachims_03a.pdf
Tags: clustering contraint spectral
Abstract: We present a new method for transductive learning, which can be seen as a transductive version of the k nearest-neighbor classifier. Unlike for many other transductive learning methods, the training problem has a meaningful relaxation that can be solved globally optimally using spectral methods. We propose an algorithm that robustly achieves good generalization performance and that can be trained efficiently. A key advantage of the algorithm is that it does not require additional heuristics to avoid unbalanced splits. Furthermore, we show a connection to transductive Support Vector Machines, and that an effective Co-Training algorithm arises as a special case.
| URL | BibTeX  
@inproceedings{citeulike:105008,
title = {Transductive Learning via Spectral Graph Partitioning},
address = {Washington DC},
author = {Thorsten Joachims},
booktitle = {Proceedings of the Twentieth International Conference on Machine Learning (ICML-2003)},
url = {http://www.cs.cornell.edu/People/tj/publications/joachims_03a.pdf},
year = {2003},
abstract = {We present a new method for transductive learning, which can be seen as a transductive version of the k nearest-neighbor classifier. Unlike for many other transductive learning methods, the training problem has a meaningful relaxation that can be solved globally optimally using spectral methods. We propose an algorithm that robustly achieves good generalization performance and that can be trained efficiently. A key advantage of the algorithm is that it does not require additional heuristics to avoid unbalanced splits. Furthermore, we show a connection to transductive Support Vector Machines, and that an effective Co-Training algorithm arises as a special case.},
priority = {4}, citeulike-article-id = {105008},
keywords = {clustering contraint spectral }
}