This page contains resources about Bayesian Model Selection and Bayesian Model Comparison.
Subfields and Concepts Edit
- Bayes Factor
- Bayesian Model Evidence / Marginal Likelihood
- Bayesian Occam's Razor
- Minimum Message Length (MML)
- Bayesian Model Averaging (in Ensemble Learning)
- Reversible jump MCMC
- Birth and death MCMC
- Bayesian Model Combination
- Type II Maximum Likelihood / Empirical Bayes / Evidence Approximation
- Type II Maximum a posteriori (MAP)
- Laplace Approximation
- Approximate Bayesian Computation (ABC)
- Akaike Information Criterion (AIC)
- Bayesian Information Criterion (BIC)
- Deviance Information Criterion (DIC)
- Bayesian Predictive Information Criterion (BPIC)
- Focused Information Criterion (FIC)
Books and Book Chapters Edit
- Theodoridis, S. (2015). "Chapter 12: Bayesian Learning" Machine Learning: A Bayesian and Optimization Perspective. Academic Press.
- Barber, D. (2012). "Chapter 12: Bayesian Model Selection". Bayesian Reasoning and Machine Learning. Cambridge University Press.
- Murphy, K. P. (2012). "Section 5.3: Bayesian model selection". Machine Learning: A Probabilistic Perspective. MIT Press.
- Claeskens, G., & Hjort, N. L. (2008). Model selection and model averaging. Cambridge: Cambridge University Press.
- Bishop, C. M. (2006). "Section 3.4: Bayesian Model Comparison". Pattern Recognition and Machine Learning. Springer.
- MacKay, D. J. (2003). "Chapter 28: Model Comparison and Occam’s Razor". Information Theory, Inference and Learning Algorithms. Cambridge University Press.
Scholarly Articles Edit
- Alston, C., Kuhnert, P., Choy, S. L., McVinish, R., & Mengersen, K. (2005). Bayesian model comparison: Review and discussion. International Statistical Insitute, 55th session.
- Wasserman, L. (2000). Bayesian model selection and model averaging. Journal of mathematical psychology, 44(1), 92-107.
- Kass, R. E., & Raftery, A. E. (1993), Bayes factors and model uncertainty, Technical Report 254, Department of Statistics, University of Washington
See also Edit
Other resources Edit
Community content is available under CC-BY-SA unless otherwise noted.