High-Dimensional Bayesian Classifiers Using Non-Local Priors

Common goals in classification problems are (i) obtaining predictions and (ii) identifying subsets of highly predictive variables. Bayesian classifiers quantify the uncertainty in all steps of the prediction. However, common Bayesian procedures can be slow in excluding features with no predictive power (Johnson & Rossell. (2010). In certain high-dimensional setups the posterior probability assigned to the correct set of predictors converges to 0 (Johnson and Rossell 2012). We study the use of non-local priors (NLP), which overcome the above mentioned limitations. We introduce a new family of NLP and derive efficient MCMC schemes.

[1]  Alex E. Lash,et al.  Gene Expression Omnibus: NCBI gene expression and hybridization array data repository , 2002, Nucleic Acids Res..

[2]  Raphael Gottardo,et al.  Markov Chain Monte Carlo With Mixtures of Mutually Singular Distributions , 2008 .

[3]  T. Yeatman,et al.  Experimentally derived metastasis gene expression profile predicts recurrence and death in patients with colon cancer. , 2010, Gastroenterology.

[4]  Guido Consonni,et al.  Moment priors for Bayesian model choice with applications to directed acyclic graphs , 2011 .

[5]  A. Raftery,et al.  Markov chain Monte Carlo with mixtures of singular distributions , 2004 .

[6]  S. Chib,et al.  Bayesian analysis of binary and polychotomous response data , 1993 .

[7]  Jianqing Fan,et al.  New Estimation and Model Selection Procedures for Semiparametric Modeling in Longitudinal Data Analysis , 2004 .

[8]  Hans Clevers,et al.  The intestinal stem cell signature identifies colorectal cancer stem cells and predicts disease relapse. , 2011, Cell stem cell.

[9]  V. Johnson,et al.  On the use of non‐local prior densities in Bayesian hypothesis tests , 2010 .

[10]  V. Johnson,et al.  Bayesian Model Selection in High-Dimensional Settings , 2012, Journal of the American Statistical Association.

[11]  T. Ørntoft,et al.  Metastasis-Associated Gene Expression Changes Predict Poor Outcomes in Patients with Dukes Stage B and C Colorectal Cancer , 2009, Clinical Cancer Research.

[12]  James G. Scott,et al.  Bayes and empirical-Bayes multiplicity adjustment in the variable-selection problem , 2010, 1011.2333.

[13]  Alan E. Gelfand,et al.  Model choice: A minimum posterior predictive loss approach , 1998, AISTATS.

[14]  T. Barrette,et al.  Oncomine 3.0: genes, pathways, and networks in a collection of 18,000 cancer gene expression profiles. , 2007, Neoplasia.