An Accurate Measure for Multilayer Perceptron Tolerance to Additive Weight Deviations

机译：对添加重量偏差的多层感知耐受性的准确度量

获取原文

页面导航

摘要
著录项
相似文献
相关主题

摘要

The inherent fault toleance of artificial neural networks (ANNs) is usually assumed, but several authors have claimed that ANNs are not always fault tolerant and have demonstrated the need to evalutate their robustness by quantitative measures. For this purpose, various alternatives have been proposed. In this paper we show the direct relation between the mean square error (MSE) and the statistical sensitivity to weight deviations, defining a measure of tolerance based on statistical sentitivity that we have called Mean Square Sensitivity (MSS); this allows us to predict accurately the degradation of the MSE when the weight values change and so constitutes a useful parameter for choosing between different configurations of MLPs. The experimental results obtained for different MLPs are shown and demonstrate the validity of our model.

机译：通常假设人工神经网络（ANNS）的固有故障占性，但是几位作者声称ANNS并不总是容忍，并且已经证明需要通过定量措施来评估其鲁棒性。为此目的，已经提出了各种替代方案。在本文中，我们展示了平均方误差（MSE）与重量偏差的统计敏感性之间的直接关系，根据我们称为均值平方敏感度（MSS）的统计周值来定义耐受性的量度;这允许我们准确地预测MSE时的MSE的劣化，因此构成用于在MLP的不同配置之间选择的有用参数。显示了对不同MLP获得的实验结果显示并展示了我们模型的有效性。

著录项

来源
《International work-conference on artificial and natural neural networks》|1999年||共10页
会议地点
作者
Jose L.Bernier; J.Ortega; M.M.Rodriguez; I.Rojas; A.Prieto;
展开▼
作者单位

展开▼
会议组织
原文格式 PDF
正文语种
中图分类自动化系统理论;
关键词

相似文献

外文文献
中文文献
专利

1. Improving the tolerance of multilayer perceptrons by minimizing the statistical sensitivity to weight deviations [J] . J.L.Bernier, J.Ortega, I.Rojas Neurocomputing . 2000,第1a4期

机译：通过最小化对重量偏差的统计敏感性来提高多层感知器的容忍度
2. Synaptic weight noise during multilayer perceptron training: fault tolerance and training improvements [J] . Murray A.F., Edwards P.J. IEEE Transactions on Neural Networks . 1993,第4期

机译：多层感知器训练过程中的突触重量噪声：容错和训练改进
3. Minimizing Hexapod Robot Foot Deviations Using Multilayer Perceptron: [J] . Vytautas Valaitis, Tomas Luneckas, Mindaugas Luneckas, International Journal of Advanced Robotic Systems . 2015,第12期

机译：使用多层感知器最大程度地减少六足机器人的脚部偏差：
4. An Accurate Measure for Multilayer Perceptron Tolerance to Additive Weight Deviations [C] . Jose L.Bernier, J.Ortega, M.M.Rodriguez, Engineering applications of bio-inspired artificial neural networks . 1999

机译：多层感知器对加性重量偏差的容忍度的精确度量
5. Iterative feature weighting for identification of relevant features in machine learning: With multilayer perceptron, radial basis function and support vector architectures. [D] . Duan, Baofu. 2005

机译：用于识别机器学习中相关特征的迭代特征加权：具有多层感知器，径向基函数和支持向量体系结构。
6. A deep learning-based method for grip strength prediction: Comparison of multilayer perceptron and polynomial regression approaches [O] . Jaejin Hwang, Jinwon Lee, Kyung-Sun Lee 2021

机译：基于深度学习的抓取强度预测方法：多层感知和多项式回归方法的比较
7. Minimizing Hexapod Robot Foot Deviations Using Multilayer Perceptron [O] . Vytautas Valaitis, Tomas Luneckas, Mindaugas Luneckas, 2015

机译：使用多层感知器最小化六足机器人脚部偏差
8. Specifications and Tolerances for Reference Standards and Field Standard Weights and Measures. 1. Specifications and Tolerances for Field Standard Weights (NIST) Class F. Revised 1990 [R] . Harris, G. L. 1990

机译：参考标准和现场标准重量和测量的规格和公差。 1.现场标准重量（NIsT）F级的规格和公差。修订于1990年

An Accurate Measure for Multilayer Perceptron Tolerance to Additive Weight Deviations

摘要

著录项

相似文献

相关主题

期刊订阅