图 书 馆
在错误的MDL贝叶斯分类的次优性Suboptimality of MDL and Bayes in Classification under Misspecification |
|
| 课程网址: | http://videolectures.net/mcslw04_grunwald_smbcu/ |
| 主讲教师: | Peter Grünwald |
| 开课单位: | 数学与计算机科学中心 |
| 开课时间: | 2007-02-25 |
| 课程语种: | 英语 |
| 中文简介: |  |
| 课程简介: | We show that forms of Bayesian and MDL learning that are often applied to classification problems can be *statistically inconsistent*. We present a large family of classifiers and a distribution such that the best classifier within the model has generalization error (expected 0/1-prediction loss) almost 0. Nevertheless, no matter how many data are observed, both the classifier inferred by MDL and the classifier based on the Bayesian posterior will behave much worse than this best classifier in the sense that their expected 0/1-prediction loss is substantially larger. Our result can be re-interpreted as showing that under misspecification, Bayes and MDL do not always converge to the distribution in the model that is closest in KL divergence to the data generating distribution. We compare this result with earlier results on Bayesian inconsistency by Diaconis, Freedman and Barron. |
| 关 键 词: | 贝叶斯; MDL学习; 模型泛化误差; 数据生成分布 |
| 课程来源: | 视频讲座网 |
| 最后编审: | 2020-07-29：yumf |
| 阅读次数: | 58 |