Pattern recognition with a Bayesian kernel combination machine
Pattern recognition with a Bayesian kernel combination machine
复制标题
DOI：
10.1016/j.patrec.2008.08.016
复制
发表时间：
2009-01-01
影响因子：
5.1
通讯作者：
Girolami, Mark A.
中科院分区：
文献类型：
--
作者：
Damoulas, Theodoros;Girolami, Mark A.
本文描述了一种多类问题的贝叶斯分类方法，该方法将多种信息来源和多个特征空间进行信息结合。该方法基于核方法的最新进展，其中通过核组合实现多个目标描述符或特征空间的集成。每个核构造一个特定特征空间中对象之间的相似度度量，然后有一个跨模态的公共度量，就可以构造一个整体组合。我们遵循分层贝叶斯方法，该方法引入了随机变量的先验分布，并构造了吉布斯抽样马尔可夫链蒙特卡罗（MCMC）解，该解自然地从所采用的多项概率似然中推导出来。该方法是可能的确定性近似（如变分或最大后验估计）的基础，并与手写数字分类的知名分类器组合方法进行了比较。该方法的结果表明，与最佳单个分类器相比，该方法有了显著的改进，并且与最佳多个分类器组合的性能相匹配，同时减少了组合分类器的计算需求，并提供了有关贡献源重要性的额外信息。(C) 2008 Elsevier B.V.版权所有
In this paper, we describe a Bayesian classification method that informatively combines diverse sources of information and Multiple feature spaces for multiclasss problems. The proposed method is based on recent advances in kernel approaches where the integration of multiple object descriptors, or feature spaces, is achieved via kernel combination. Each kernel constructs a similarity metric between objects in a particular feature space and then having a common metric across modalities an overall combination can be constructed. We follow a hierarchical Bayesian approach, which introduces prior distributions over random variables and we construct a Gibbs sampling Markov chain Monte Carlo (MCMC) solution which is naturally derived from the employed multinomial probit likelihood. The methodology is the basis for possible deterministic approximations Such as variational or maximum-a-posteriori estimators, and it is compared against the well-known classifier combination methods on the classification of handwritten numerals. The results of the proposed method show a significant improvement over the best individual classifier and match the performance of the best Multiple classifier combination, whilst reducing the Computational requirements of combining classifiers and offering additional information on the significance of the contributing Sources. (C) 2008 Elsevier B.V. All rights reserved.