Bayesian Learning of Measurement and Structural Models Export

@inproceedings{citeulike:6043825, abstract = {We present a Bayesian search algorithm for learning the structure of latent variable models of continuous variables. We stress the importance of applying search operators designed especially for the parametric family used in our models. This is performed by searching for subsets of the observed variables whose covariance matrix can be represented as a sum of a matrix of low rank and a diagonal matrix of residuals. The resulting search procedure is relatively efficient, since the main search operator has a branch factor that grows linearly with the number of variables. The resulting models are often simpler and give a better fit than models based on generalizations of factor analysis or those derived from standard hill-climbing methods.}, address = {New York, NY, USA}, author = {Silva, Ricardo and Scheines, Richard}, booktitle = {Proceedings of the 23rd International Conference on Machine Learning (ICML '06)}, citeulike-article-id = {6043825}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1143948}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/1143844.1143948}, doi = {10.1145/1143844.1143948}, isbn = {1-59593-383-2}, keywords = {bayesian, graphical\_models}, location = {Pittsburgh, Pennsylvania}, pages = {825--832}, posted-at = {2009-10-30 18:26:13}, priority = {2}, publisher = {ACM}, title = {Bayesian Learning of Measurement and Structural Models}, url = {http://dx.doi.org/10.1145/1143844.1143948}, year = {2006} }

TY - CONF ID - citeulike:6043825 L3 - citeulike-article-id:6043825 N2 - We present a Bayesian search algorithm for learning the structure of latent variable models of continuous variables. We stress the importance of applying search operators designed especially for the parametric family used in our models. This is performed by searching for subsets of the observed variables whose covariance matrix can be represented as a sum of a matrix of low rank and a diagonal matrix of residuals. The resulting search procedure is relatively efficient, since the main search operator has a branch factor that grows linearly with the number of variables. The resulting models are often simpler and give a better fit than models based on generalizations of factor analysis or those derived from standard hill-climbing methods. CY - Pittsburgh, Pennsylvania SN - 1-59593-383-2 AD - New York, NY, USA EP - 832 TI - Bayesian Learning of Measurement and Structural Models PB - ACM T2 - Proceedings of the 23rd International Conference on Machine Learning (ICML '06) SP - 825 KW - bayesian KW - graphical_models AU - Silva, Ricardo AU - Scheines, Richard PY - 2006/// UR - http://dx.doi.org/10.1145/1143844.1143948 ER -