【机器学习】5 Bayesian statistics

本章目录

5 Bayesian statistics 149
5.1 Introduction 149
5.2 Summarizing posterior distributions 149
5.2.1 MAP estimation 149
5.2.2 Credible intervals 152
5.2.3 Inference for a difference in proportions 154
5.3 Bayesian model selection 155
5.3.1 Bayesian Occam’s razor 156
5.3.2 Computing the marginal likelihood (evidence) 158
5.3.3 Bayes factors 163
5.3.4 Jeffreys-Lindley paradox * 164
5.4 Priors 165
5.4.1 Uninformative priors 165
5.4.2 Jeffreys priors * 166
5.4.3 Robust priors 168
5.4.4 Mixtures of conjugate priors 168
5.5 Hierarchical Bayes 171
5.5.1 Example: modeling related cancer rates 171
5.6 Empirical Bayes 172
5.6.1 Example: beta-binomial model 173
5.6.2 Example: Gaussian-Gaussian model 173
5.7 Bayesian decision theory 176
5.7.1 Bayes estimators for common loss functions 177
5.7.2 The false positive vs false negative tradeoff 180
5.7.3 Other topics * 184

github下载链接：https://github.com/916718212/Machine-Learning-A-Probabilistic-Perspective-.git