arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1609.05486cs.LGstat.ML
收藏

概率特征选择与分类向量机

Probabilistic Feature Selection and Classification Vector Machine

University of Science and Technology of China(中国科学技术大学);University of Amsterdam(阿姆斯特丹大学)

Bingbing Jiang, Chang Li, Maarten de Rijke, Xin Yao, Huanhuan Chen

摘要:

稀疏贝叶斯学习是一种最先进的监督学习算法,它可以从输入数据中选择相关样本的子集,并进行可靠的概率预测。然而,在高维数据与不相关的功能,传统的稀疏贝叶斯分类器的性能下降和效率低,无法消除不相关的功能。为了解决这个问题,我们提出了一种新的稀疏贝叶斯嵌入式特征选择方法,采用截断高斯分布作为样本和特征先验。所提出的方法,称为概率特征选择和分类向量机(PFCVMLP),能够同时选择相关的特征和样本的分类任务。为了得到解析解,拉普拉斯近似被用来计算近似后验和边缘似然。最后,参数和超参数优化的第二类最大似然方法。在三个数据集上的实验从两个方面验证了PFCVMLP的性能:分类性能和特征选择的有效性。最后,我们分析了PFCVMLP的泛化性能,并给出了PFCVMLP的泛化误差界。通过收紧边界,证明了特征选择的重要性。

英文摘要:

Sparse Bayesian learning is a state-of-the-art supervised learning algorithm that can choose a subset of relevant samples from the input data and make reliable probabilistic predictions. However, in the presence of high-dimensional data with irrelevant features, traditional sparse Bayesian classifiers suffer from performance degradation and low efficiency by failing to eliminate irrelevant features. To tackle this problem, we propose a novel sparse Bayesian embedded feature selection method that adopts truncated Gaussian distributions as both sample and feature priors. The proposed method, called probabilistic feature selection and classification vector machine (PFCVMLP ), is able to simultaneously select relevant features and samples for classification tasks. In order to derive the analytical solutions, Laplace approximation is applied to compute approximate posteriors and marginal likelihoods. Finally, parameters and hyperparameters are optimized by the type-II maximum likelihood method. Experiments on three datasets validate the performance of PFCVMLP along two dimensions: classification performance and effectiveness for feature selection. Finally, we analyze the generalization performance and derive a generalization error bound for PFCVMLP . By tightening the bound, the importance of feature selection is demonstrated.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑