Combining active learning and semi-supervised learning using gaussian fields and harmonic functions

@misc{citeulike:1877838, abstract = {Active and semi-supervised learning are important techniques when labeled data are scarce. We combine the two under a Gaussian random field model. Labeled and unlabeled data are represented as vertices in a weighted graph, with edge weights encoding the similarity between instances. The semi-supervised learning problem is then formulated in terms of a Gaussian random field on this graph, the mean of which is characterized in terms of harmonic functions. Active learning is performed on top of...}, author = {Zhu, X. and Lafferty, J. and Ghahramani, Z. }, citeulike-article-id = {1877838}, keywords = {active, active\_learning, fields, functions, gaussian, harmonic, learning, semi-supervised}, posted-at = {2007-11-07 15:19:00}, priority = {0}, title = {Combining active learning and semi-supervised learning using gaussian fields and harmonic functions}, url = {http://citeseer.ist.psu.edu/681419.html}, year = {2003} }

TY - GEN ID - citeulike:1877838 L3 - citeulike-article-id:1877838 TI - Combining active learning and semi-supervised learning using gaussian fields and harmonic functions N2 - Active and semi-supervised learning are important techniques when labeled data are scarce. We combine the two under a Gaussian random field model. Labeled and unlabeled data are represented as vertices in a weighted graph, with edge weights encoding the similarity between instances. The semi-supervised learning problem is then formulated in terms of a Gaussian random field on this graph, the mean of which is characterized in terms of harmonic functions. Active learning is performed on top of... KW - active KW - active_learning KW - fields KW - functions KW - gaussian KW - harmonic KW - learning KW - semi-supervised AU - Zhu, X AU - Lafferty, J AU - Ghahramani, Z PY - 2003/// UR - http://citeseer.ist.psu.edu/681419.html ER -