• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文学位 >Reduced HyperBF networks: Practical optimization, regularization, and applications in bioinformatics.
【24h】

Reduced HyperBF networks: Practical optimization, regularization, and applications in bioinformatics.

机译：简化的HyperBF网络：实用的优化，正则化及其在生物信息学中的应用。

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

A hyper basis function network (HyperBF) is a generalized radial basis function network (RBF) where the activation function is a radial function of a weighted distance. The local weighting of the distance accounts for the variation in local scaling and discriminative power along each feature. Such generalization makes HyperBF networks capable of interpolating decision functions with high accuracy. However, such complexity makes HyperBF networks susceptible to overfitting. Moreover, training a HyperBF network demands weights, centers and local scaling factors to be optimized simultaneously. In the case of a relatively large dataset with a large network structure, such optimization becomes computationally challenging.In addition, a practical training approach for constructing HyperBF networks is presented. This approach uses hierarchal clustering to initialize neurons followed by a gradient optimization using a scaled Rprop algorithm with a localized partial backtracking step (iSRprop). Experimental results on a number of datasets show a faster and smoother convergence than the regular Rprop algorithm.The proposed Reduced HyperBF network is applied to two problems in bioinformatics. The first is the detection of transcription start sites (TSS) in human DNA. A novel method for improving the accuracy of TSS recognition for recently published methods is proposed. This method incorporates a new metric feature based on oligonucleotide positional frequencies.The second application is the accurate classification of microarray samples. A new feature selection algorithm based on a Reduced HyperBF network is proposed. The method is applied to two microarray datasets and is shown to select a minimal subset of features with high discriminative information. The algorithm is compared to two widely used methods and is shown to provide competitive results.In this work, a new regularization method that performs soft local dimension reduction and weight decay is presented. The regularized HyperBF (Reduced HyperBF) network is shown to provide classification accuracy comparable to a Support Vector Machines (SVM) while requiring a significantly smaller network structure. Furthermore, the soft local dimension reduction is shown to be informative for ranking features based on their localized discriminative power.In both applications, the final Reduced HyperBF network is used for higher level analysis. Significant neurons can indicate subpopulations, while local active features provide insight into the characteristics of the subpopulation in specific and the whole class in general.
机译：超基函数网络（HyperBF）是广义径向基函数网络（RBF），其中激活函数是加权距离的径向函数。距离的局部权重说明沿每个要素的局部缩放比例和判别力的变化。这种概括使HyperBF网络能够高精度地插入决策函数。但是，这种复杂性使HyperBF网络易于过度拟合。此外，训练HyperBF网络需要同时优化权重，中心和局部缩放因子。在具有较大网络结构的相对较大的数据集的情况下，这种优化在计算上具有挑战性。此外，提出了一种用于构建HyperBF网络的实用训练方法。该方法使用层次聚类来初始化神经元，然后使用带有局部局部回溯步骤（iSRprop）的缩放Rprop算法进行梯度优化。在大量数据集上的实验结果表明，与常规的Rprop算法相比，其收敛速度更快，更平滑。拟议的简化HyperBF网络被应用于生物信息学中的两个问题。首先是检测人类DNA中的转录起始位点（TSS）。针对最近发表的方法，提出了一种提高TSS识别准确性的新方法。该方法结合了基于寡核苷酸位置频率的新度量标准功能。第二个应用是微阵列样品的准确分类。提出了一种基于简化HyperBF网络的特征选择算法。该方法应用于两个微阵列数据集，并显示为选择具有高区分性信息的特征的最小子集。该算法与两种广泛使用的方法进行了比较，并显示出具有竞争性的结果。在这项工作中，提出了一种新的进行软局部尺寸减小和权重衰减的正则化方法。所示的正则化HyperBF（精简HyperBF）网络可提供与支持向量机（SVM）相当的分类精度，同时所需的网络结构要小得多。此外，软局部维数缩减显示出可根据特征的本地化区分能力对特征进行排序。在这两种应用中，最终的简化HyperBF网络都用于更高级别的分析。重要的神经元可以指示亚群，而局部活动特征可以洞察特定类别和整个类别中亚群的特征。

著录项

  • 作者

    Mahdi, Rami Nezar.;

  • 作者单位

    University of Louisville.;

  • 授予单位 University of Louisville.;
  • 学科 Biology Bioinformatics.Artificial Intelligence.Computer Science.
  • 学位 Ph.D.
  • 年度 2010
  • 页码 119 p.
  • 总页数 119
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号