Semi-supervised learning with trees Export

@article{citeulike:1714729, abstract = {We describe a nonparametric Bayesian approach to generalizing from few labeled examples, guided by a larger set of unlabeled objects and the assumption of a latent tree-structure to the domain. The tree (or a distribution over trees) may be inferred using the unlabeled data. A prior over concepts generated by a mutation process on the inferred tree(s) allows efficient computation of the optimal Bayesian classification function from the labeled examples. We test our approach on eight real-world datasets.}, author = {Kemp, C. and Griffiths, T. and Stromsten, S. and Tenenbaum, J.}, citeulike-article-id = {1714729}, journal = {Advances in Neural Information Processing Systems}, keywords = {causal\_relation, induction, learning}, posted-at = {2007-10-01 11:06:24}, priority = {4}, title = {Semi-supervised learning with trees}, year = {2004} }

TY - JOUR ID - citeulike:1714729 L3 - citeulike-article-id:1714729 N2 - We describe a nonparametric Bayesian approach to generalizing from few labeled examples, guided by a larger set of unlabeled objects and the assumption of a latent tree-structure to the domain. The tree (or a distribution over trees) may be inferred using the unlabeled data. A prior over concepts generated by a mutation process on the inferred tree(s) allows efficient computation of the optimal Bayesian classification function from the labeled examples. We test our approach on eight real-world datasets. JF - Advances in Neural Information Processing Systems TI - Semi-supervised learning with trees KW - causal_relation KW - induction KW - learning AU - Kemp, C AU - Griffiths, T AU - Stromsten, S AU - Tenenbaum, J PY - 2004/// ER -