Learning Bounds of ERM Principle for Sequences of Time-Dependent Samples
Learning Bounds of ERM Principle for Sequences of Time-Dependent Samples
复制标题
DOI：
10.1155/2015/826812
复制
发表时间：
2015-11
影响因子：
1.4
通讯作者：
M. Yao;Chao Zhang;Wei Wu
中科院分区：
文献类型：
--
作者：
M. Yao;Chao Zhang;Wei Wu
学习理论中的许多泛化结果都是在样本独立且同分布的假设下建立起来的。然而，在实际应用中，许多学习任务都涉及到与时间相关的数据。在本文中，我们提出了一个理论框架来分析经验风险最小化（ERM）原则对时间相关样本序列（TDS）的泛化性能。特别地，我们首先给出了TDS的ERM原理的泛化界。通过引入一些辅助量，进一步分析了TDS的ERM原理的泛化性质和渐近行为。
Many generalization results in learning theory are established under the assumption that samples are independent and identically distributed (i.i.d.). However, numerous learning tasks in practical applications involve the time-dependent data. In this paper, we propose a theoretical framework to analyze the generalization performance of the empirical risk minimization (ERM) principle for sequences of time-dependent samples (TDS). In particular, we first present the generalization bound of ERM principle for TDS. By introducing some auxiliary quantities, we also give a further analysis of the generalization properties and the asymptotical behaviors of ERM principle for TDS.