Learning coefficients and reproducing true probability functions in learning systems
Learning coefficients and reproducing true probability functions in learning systems
复制标题
学习系统中的学习系数和再现真实概率函数
DOI：
10.1007/978-3-319-48812-7_44
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
Miki Aoyagi
中科院分区：
文献类型：
--
作者：
三谷健一;斎藤吉助;高橋泰嗣;Tomoyuki Tanigawa;Miki Aoyagi
近年来，广泛应用的信息准则（WAIC）模型选择方法被用于从学习系统的数据中再现和估计概率函数。贝叶斯估计中的学习系数用于衡量奇异学习模型的学习效率，在WAIC方法中具有重要作用。数学上，学习系数是相对熵的对数标准阈值。本文考虑了统计学习理论中的Vandermonde矩阵型奇异学习系数。
Recently, the widely applicable information criterion (WAIC) model selection method has been considered for reproducing and estimating a probability function from data in a learning system. The learning coefficient in Bayesian estimation serves to measure the learning efficiency in singular learning models, and has an important role in the WAIC method. Mathematically, the learning coefficient is the log canonical threshold of the relative entropy. In this paper, we consider the Vandermonde matrix-type singularity learning coefficients in statistical learning theory.