×

Remark on membership functions in neuro-fuzzy systems. (English) Zbl 1189.68102

Cyran, Krzysztof A. (ed.) et al., Man-machine interactions. Papers based on the presentations at the international conference (ICMMI 2009), Kocierz Pass, Poland, September 25–27, 2009. Berlin: Springer (ISBN 978-3-642-00562-6/pbk; 978-3-642-00563-3/ebook). Advances in Intelligent and Soft Computing 59, 291-297 (2009).
Summary: The sigmoidal membership function applied in the neuro-fuzzy systems with hierarchical input domain partition and gradient tuning method may deteriorate the tuning process. The function’s high value plateau with very low derivative’s value stops the gradient based tuning procedure. This leads to less adequate models and poorer results elaborated by the system. In such systems the membership function should satisfy the condition that the low values of derivatives in respect of the function parameters should be followed by the low values of membership function itself. The points of the domain that do not fulfil this condition may only be isolated point of the domain. The function should have no high membership plateaux. The functions suitable for systems with hierarchical input domain partition are bell-like functions as Gaussian, generalised bell function, (a)symmetric \(\pi \) function.
For the entire collection see [Zbl 1181.68003].

MSC:

68T05 Learning and adaptive systems in artificial intelligence
PDFBibTeX XMLCite
Full Text: DOI