Asymptotically minimax regret by Bayes mixtures

Asymptotically minimax regret by Bayes mixtures
复制标题

贝叶斯混合的渐近最小最大遗憾

DOI:
10.1109/isit.1998.708923
复制
发表时间:
1998
期刊:
Proceedings. 1998 IEEE International Symposium on Information Theory (Cat. No.98CH36252)
影响因子:
--
通讯作者:
Andrew R. Barron
Andrew R. Barron
中科院分区:
--
文献类型:
--
作者:
Jun'ichi Takeuchi;Andrew R. Barron

文献摘要

被引文献

相似文献

研究了序列x/sup n/ = x/sub 1/x/sub 2/…的数据压缩、赌博和预测问题。x/sub n/从一定的字母表X,在遗憾(Shtarkov 1988)和冗余方面的一般指数家庭,一般光滑的家庭,也马尔可夫源。特别是,我们证明了Jeffreys混合物的变体渐近达到其极大极小值。
We study the problem of data compression, gambling and prediction of a sequence x/sup n/ = x/sub 1/x/sub 2/...x/sub n/ from a certain alphabet X, in terms of regret (Shtarkov 1988) and redundancy with respect to a general exponential family, a general smooth family, and also Markov sources. In particular, we show that variants of Jeffreys mixture asymptotically achieve their minimax values.