Ricardo@UCL
Ricardo Silva
Department of Statistical Science, UCL
Gower Street - London - WC1E 6BT
University College London
ricardo [-at-] stats [dot] ucl [dot] ac [dot] uk

Fast facts:
I am a Lecturer in Statistics in UCL working on computational approaches for graphical latent variable models, relational models and causal inference.
 

Departmental Tutor office hours:
Usually Mondays 11am-noon, Wednesdays noon-1pm, Fridays 2pm-3pm. In general I will not be available at other times without an appointment. My office is number 139 at the 1-19 Torrington Place building (corner of Torrington Place with Tottenham Court Road).
 


CV

Software
 
  • WPP: Causal inference through a Witness Protection Program. A method for inferring bounds on average causal effects based on a combination of linear programming and Bayesian inference. R code.

  • XCOP: Dirichlet process mixture of tree-structured copulas. MCMC method for Bayesian inference with multivariate copulas and missing data. R/C++ code.

  • GPSEM: Gaussian process structural equation models with latent variables. Samples latent variables and parameters from the posterior distribution. MATLAB/C++ code.

  • XGP: a new Gaussian process relational classifier. A new family of transductive classifiers using relational information. MATLAB code.

  • dmgBayes: software for Bayesian inference in mixed graph models. Basic functionality for Gaussian modeling with mixed graph models. Java code.

  • RankSearch: Bayesian structure learning for latent variable models. Sparse factor model learning with dependent latent variables. Java code.

  • BuildPureClusters: learning to measure hidden common causes. An approach for building measurement models of hidden common causes from data without specifying the hidden common causes a priori. Part of the Tetrad project.

  • Publications
     
  • Silva, R. and Evans, R. (2014). Causal inference through a witness protection program Advances in Neural Information Processing Systems, NIPS. [Code] [arXiv]

  • Silva, R. (2014). Causality In G. Webb and C. Sammut, eds., Springer Encyclopedia of Machine Learning and Data Mining, to appear (this is an update of the 2010 article below). [Tutorial@Imperial College]

  • Silva, R. (2014). Bayesian inference in cumulative distribution fields Springer Proceedings in Mathematics & Statistics, 12th Brazilian Meeting on Bayesian Statistics, to appear.

  • Kalaitzis, A. and Silva, R. (2013). Flexible sampling of discrete data correlations without the marginal distributions Advances in Neural Information Processing Systems, NIPS. [Supplementary Material]

  • Silva, R. (2013). A MCMC approach for learning the structure of Gaussian acyclic directed mixed graphs In Giudici, Ingrassia and Vichi, eds., Statistical Models for Data Analysis. Springer. [Draft] [Code]

  • Sanborn, A. and Silva, R. (2013). Constraining bridges between levels of analysis: A computational justification for Locally Bayesian Learning. Journal of Mathematical Psychology, 57, 94-106. [Manuscript]

  • Silva, R. (2012). Latent composite likelihood learning for the structured canonical correlation model . Proceedings of the 28th Conference on Uncertainty in Artificial Intelligence, UAI 2012. [Supplementary Material] [Code] [Poster]

  • Silva, R. (2011). Thinning measurement models and questionnaire design . Advances in Neural Information Processing Systems 24, NIPS 2011. [Supplementary Material] [Code]

  • Zhang, J. and Silva, R. (2011). Discussion of ``Learning Equivalence Classes of Acyclic Models with Latent and Selection Variables from Multiple Datasets with Overlapping Variables.''. Proceedings of the 14th International Conference on Artificial Intelligence and Statistics, AISTATS 2011. [Slides]

  • Silva, R; Blundell, C. and Teh, Y.W. (2011). Mixed cumulative distribution networks . Proceedings of the 14th International Conference on Artificial Intelligence and Statistics, AISTATS 2011. [Supplementary Material] [Code (Coming Soon)] [Slides]

  • Silva, R. (2010). Causality. In Encyclopedia of Machine Learning. Claude Sammut, ed. Springer-Verlag. ISBN: 978-0387-30768-8

  • Silva, R. and Gramacy, R. (2010). Gaussian process structural equation models with latent variables . Proceedings of the 26th Conference on Uncertainty on Artificial Intelligence, UAI 2010. [Code] [Slides]

  • Silva, R. (2010). Measuring latent causal structure. P. McKay Illari, F. Russo and J. Williamson (eds.), Causality in the Sciences, to appear. [Draft]

  • Silva, R.; Heller, K.; Ghahramani, Z. and Airoldi, E. (2010). Ranking relations using analogies in biological and information networks. Annals of Applied Statistics, to appear. [arXiv]

  • Silva, R. and Ghahramani, Z. (2009). The hidden life of latent variables: Bayesian learning with mixed graph models. Journal of Machine Learning Research 10, 1187--1238. [Code]

  • Sanborn, A. N. and Silva, R. (2009). Belief propagation and locally Bayesian learning. 31st Annual Conference of the Cognitive Science Society.

  • Silva, R. and Gramacy, R. (2009). MCMC methods for Bayesian mixtures of copulas. Artificial Intelligence and Statistics, AISTATS 2009. [Code]

  • Silva, R. and Ghahramani, Z. (2009). Factorial mixture of Gaussians and the marginal independence model. Artificial Intelligence and Statistics, AISTATS 2009. [Code]

  • Silva, R.; Chu, W. and Ghahramani, Z. (2007). Hidden common cause relations in relational learning. Neural Information Processing Systems, NIPS 2007. [Code] [Data] [Poster]

  • Silva, R.; Heller, K. and Ghahramani, Z. (2007). Analogical reasoning with relational Bayesian sets. 11th International Conference on Artificial Intelligence and Statistics, AISTATS 2007.

  • Silva, R. and Scheines, R. (2006). Towards association rules with hidden variables. Proceedings of the 10th European Conference on Principles and Practice of Knowledge Discovery in Databases, PKDD 2006.

  • Silva, R. and Ghahramani, Z. (2006). Bayesian inference for Gaussian mixed graph models. Proceedings of the 22nd Conference on Uncertainty on Artificial Intelligence, UAI 2006. [Code]

  • Silva, R. and Scheines, R. (2006). Bayesian learning of measurement and structural models. Proceedings of the 23rd International Conference on Machine Learning, ICML 2006. [Code]

  • Silva, R.; Scheines, R.; Glymour, C. and Spirtes, P. (2006). Learning the structure of linear latent variable models. Journal of Machine Learning Research 7(Feb):191--246, 2006. [Code]

  • Silva, R. and Scheines, R. (2005). New d-separation identification results for learning continuous latent variable models. Proceedings of the International Conference in Machine Learning, ICML 05. Tech report version.

  • Silva, R.; Zhang, J. and Shanahan, J. G. (2005). Probabilistic workflow mining. Proceedings of Knowledge Discovery and Data Mining, KDD 05. Tech report version.

  • Silva, R. (2005). Automatic Discovery of Latent Variable Models . PhD Thesis. Machine Learning Department, Carnegie Mellon University.

  • Silva, R.; Scheines, R.; Glymour, C. and Spirtes P. (2003) "Learning measurement models for unobserved variables". Proceedings of the 19th Conference on Uncertainty on Artificial Intelligence.

  • Moody, J.; Silva, R.; Vanderwaart, J.; Ramsey, J. and Glymour, C. (2002). Classification and filtering of spectra: A case study in mineralogy. Intelligent Data Analysis 6 (6), 517-530.

  • Moody, J.; Silva, R.; Vanderwaart, J. and Glymour, C. (2001). "Data filtering for automatic classification of rocks from reflectance spectra". Proceedings of the 7th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, p. 347-352. ACM Press, San Francisco, CA. 

  • Silva, R. B. A. and Ludermir, T. B. (2001). “Hybrid systems of local basis functions”. Intelligent Data Analysis 5 (3), 227-244

  • Silva, R. B. A. and Ludermir, T. B. (2000). “Obtaining simplified rules by hybrid learning”. Proceedings of the 17th International Conference on Machine Learning, 879-886. Morgan Kaufmann, San Francisco, CA
  • Other publications/reports



    Last modification: June 07th, 2011