Exploiting informative priors for Bayesian classification and regression trees

Authors:
Nicos Angelopoulos;James Cussens
Affiliations:
Department of Computer Science, University of York, Heslington, York, UK;Department of Computer Science, University of York, Heslington, York, UK
Venue:
IJCAI'05 Proceedings of the 19th international joint conference on Artificial intelligence
Year:
2005

Citing 3
Cited 3

Convergence rates for Markov chains

SIAM Review
Stochastic Logic Programs: Sampling, Inference and Applications

UAI '00 Proceedings of the 16th Conference on Uncertainty in Artificial Intelligence
Markov Chain Monte Carlo using Tree-Based Priors on Model Structure

UAI '01 Proceedings of the 17th Conference in Uncertainty in Artificial Intelligence

Tempering for Bayesian C&RT

ICML '05 Proceedings of the 22nd international conference on Machine learning
Bayes optimal classification for decision trees

Proceedings of the 25th international conference on Machine learning
Optimal constraint-based decision tree induction from itemset lattices

Data Mining and Knowledge Discovery

Quantified Score

Hi-index	0.00

Visualization

Abstract

A general method for defining informative priors on statistical models is presented and applied specifically to the space of classification and regression trees. A Bayesian approach to learning such models from data is taken, with the Metropolis-Hastings algorithm being used to approximately sample from the posterior. By only using proposal distributions closely tied to the prior, acceptance probabilities are easily computable via marginal likelihood ratios, whatever the prior used. Our approach is empirically tested by varying (i) the data, (ii) the prior and (iii) the proposal distribution. A comparison with related work is given.