Canonical dependency analysis based on squared-loss mutual information
Canonical dependency analysis based on squared-loss mutual information
复制标题
DOI：
10.1016/j.neunet.2012.06.009
复制
发表时间：
2012-10-01
期刊：
影响因子：
7.8
通讯作者：
Sugiyama, Masashi
中科院分区：
文献类型：
--
作者：
Karasuyama, Masayuki;Sugiyama, Masashi
典型相关分析（CCA）是一种经典的降维技术，用于两组变量，迭代地找到具有最大相关性的投影方向。尽管CCA在许多实际应用领域中仍然有着重要的应用，但最近的现实世界数据往往包含更复杂的非线性相关性，无法通过经典CCA正确捕获。因此，在本文中，我们提出了一个扩展的CCA，可以有效地捕捉这种复杂的非线性相关性，通过统计依赖最大化。所提出的方法，我们称之为最小二乘典型依赖分析（LSCDA），是基于互信息的平方损失变体，除了能够捕获高阶相关性之外，它还具有各种有用的特性：例如，它可以同时找到多个投影方向（即，子空间），它不涉及密度估计，并且它配备了模型选择策略。我们通过人工和真实世界的数据集上的各种实验证明了LSCDA的有用性。(C)2012爱思唯尔有限公司保留所有权利。
Canonical correlation analysis (CCA) is a classical dimensionality reduction technique for two sets of variables that iteratively finds projection directions with maximum correlation. Although CCA is still in vital use in many practical application areas, recent real-world data often contain more complicated nonlinear correlations that cannot be properly captured by classical CCA. In this paper, we thus propose an extension of CCA that can effectively capture such complicated nonlinear correlations through statistical dependency maximization. The proposed method, which we call least-squares canonical dependency analysis (LSCDA), is based on a squared-loss variant of mutual information, and it has various useful properties besides its ability to capture higher-order correlations: for example, it can simultaneously find multiple projection directions (i.e., subspaces), it does not involve density estimation, and it is equipped with a model selection strategy. We demonstrate the usefulness of LSCDA through various experiments on artificial and real-world datasets. (C) 2012 Elsevier Ltd. All rights reserved.