Document Detail

Adaptive Ho-Kashyap rules for perceptron training.
MedLine Citation:
PMID:  18276405     Owner:  NLM     Status:  In-Data-Review    
Three adaptive versions of the Ho-Kashyap perceptron training algorithm are derived based on gradient descent strategies. These adaptive Ho-Kashyap (AHK) training rules are comparable in their complexity to the LMS and perceptron training rules and are capable of adaptively forming linear discriminant surfaces that guarantee linear separability and of positioning such surfaces for maximal classification robustness. In particular, a derived version called AHK II is capable of adaptively identifying critical input vectors lying close to class boundaries in linearly separable problems. The authors extend this algorithm as AHK III, which adds the capability of fast convergence to linear discriminant surfaces which are good approximations for nonlinearly separable problems. This is achieved by a simple built-in unsupervised strategy which allows for the adaptive grading and discarding of input vectors causing nonseparability. Performance comparisons with LMS and perceptron training are presented.
M H Hassoun; J Song
Related Documents :
20221995 - Endurance performance and nocturnal hrv indices.
21085035 - Similar dynamic hyperinflation during arm and leg exercise at similar ventilation in ch...
23618675 - Influence of intense exercise on acid-base, blood gas and electrolyte status in bulls.
22889655 - Quadriceps and hamstrings coactivation during common therapeutic exercises.
16001905 - The use of real-time ultrasound feedback in teaching abdominal hollowing exercises to h...
7174215 - Facial paralysis rehabilitation: retraining selective muscle control.
Publication Detail:
Type:  Journal Article    
Journal Detail:
Title:  IEEE transactions on neural networks / a publication of the IEEE Neural Networks Council     Volume:  3     ISSN:  1045-9227     ISO Abbreviation:  IEEE Trans Neural Netw     Publication Date:  1992  
Date Detail:
Created Date:  2008-02-15     Completed Date:  -     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  101211035     Medline TA:  IEEE Trans Neural Netw     Country:  United States    
Other Details:
Languages:  eng     Pagination:  51-61     Citation Subset:  -    
Dept. of Electr. and Comput. Eng., Wayne State Univ., Detroit, MI.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine

Previous Document:  Learning and convergence analysis of neural-type structured networks.
Next Document:  Back-propagation learning in expert networks.