|
马上注册,结交更多好友,享用更多功能,让你轻松玩转社区。
您需要 登录 才可以下载或查看,没有账号?注册
x
特殊说明:这本书费了特别大的劲找到了,对信息论感兴趣的人对作者不会陌生,珍惜和善待好书的方法就是让更多的人去读它
书名:Information Theory and Statistics: A Tutorial
作者:Imre Csiszar, Paul C. Shields
出版:NOW Publisher Inc. 2004
长度:122页
1 Preliminaries 3
2 Large deviations, hypothesis testing 11
2.1 Large deviations via types 11
2.2 Hypothesis testing 16
3 I-projections 23
4 f-Divergence and contingency tables 31
5 Iterative algorithms 43
5.1 Iterative scaling 43
5.2 Alternating divergence minimization 47
5.3 The EM algorithm 55
6 Universal coding 59
6.1 Redundancy 60
6.2 Universal codes for certain classes of processes 65
7 Redundancy bounds 77
7.1 I-radius and channel capacity 78
7.2 Optimality results 85
8 Redundancy and the MDL principle 91
8.1 Codes with sublinear redundancy growth 92
8.2 The minimum description length principle 98
A Summary of process concepts 107
References 113 |
|