Learning from Multiple sources

@article{citeulike:2363626, abstract = {We consider the problem of learning accurate models from multiple sources of “nearby” data. Given distinct samples from multiple data sources and estimates of the dissimilarities between these sources, we provide a general theory of which samples should be used to learn models for each source. This theory is applicable in a broad decision-theoretic learning framework, and yields results for classification and regression generally, and for density estimation within the exponential family. A key component of our approach is the development of approximate triangle inequalities for expected loss, which may be of independent interest.}, author = {Crammer, Koby and Kearns, Michael and Wortman, Jennifer }, booktitle = {Neural Information Processing Systems}, citeulike-article-id = {2363626}, keywords = {cotraining, hierarchy}, posted-at = {2008-02-11 19:33:57}, priority = {3}, title = {Learning from Multiple sources}, url = {http://www.cis.upenn.edu/~crammer/publications/multisource\_nips07.pdf}, year = {2006} }

TY - JOUR ID - citeulike:2363626 L3 - citeulike-article-id:2363626 TI - Learning from Multiple sources T2 - Neural Information Processing Systems N2 - We consider the problem of learning accurate models from multiple sources of “nearby” data. Given distinct samples from multiple data sources and estimates of the dissimilarities between these sources, we provide a general theory of which samples should be used to learn models for each source. This theory is applicable in a broad decision-theoretic learning framework, and yields results for classification and regression generally, and for density estimation within the exponential family. A key component of our approach is the development of approximate triangle inequalities for expected loss, which may be of independent interest. KW - cotraining KW - hierarchy AU - Crammer, Koby AU - Kearns, Michael AU - Wortman, Jennifer PY - 2006/// UR - http://www.cis.upenn.edu/~crammer/publications/multisource_nips07.pdf ER -