خط مشی دسترسیدرباره ما
ثبت نامثبت نام
راهنماراهنما
فارسی
ورودورود
صفحه اصلیصفحه اصلی
جستجوی مدارک
تمام متن
منابع دیجیتالی
رکورد قبلیرکورد بعدی
Document Type : Latin Dissertation
Language of Document : English
Record Number : 149801
Doc. No : ET21593
Main Entry : Sampath Yerramalla
Title Proper : Stability Monitoring and Analysis of Online Learning Neural Networks
Note : This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract : On-line adaptation using soft-computational learning methods is on the rise for use insafety-critical applications such as fault-tolerant flight control, maintenance of distributednetworks, implementation of high security devices, etc. The inapplicability of traditionalanalysis methods is limiting the wider use of soft-computational learning methods in safetycriticalapplications that involve online adaptation. The focus of the research is thedevelopment of non-conventional analysis techniques for the testing, verification, validationand analysis of adaptive learning components such as the online learning neural networks.Our research considers stability of online adaptation as a heuristic measure ofcorrectness in the operation of the adaptive component. The approach is based on theprinciples of stability according to Lyapunov theory, deriving mathematical stability proofs toassure convergence in neural network learning within a bounded amount of time. The analysisis applied to online learning neural networks such as the Dynamic Cell Structures, Sigma-Pi,and Adaline. This approach is applicable for learning from stationary, non-varying data. Fortime-varying training data sets, we developed the online stability monitoring methodology.Stability monitors analyze the neural networkبs learning in real-time. ROC curves present theperformance of the developed stability monitoring system as a trade off between the selectivityand sensitivity of the stability detection. Further, we derived a convergence predictionmethodology that, given the amplitude of the disturbance, predicts the number of learningtested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject : Electericl tess
: برق
electronic file name : TL44768.pdf
Title and statement of responsibility and : Stability Monitoring and Analysis of Online Learning Neural Networks [Thesis]
 
 
 
(در صورت عدم وضوح تصویر اینجا را کلیک نمایید)