----------------------------------------------------------------------
ARTIFICIAL INTELLIGENCE RESEARCH SEMINAR
Department of Computer Science
Worcester Polytechnic Institute
SPEAKER: Prof. Michael Gennert
TITLE: "Optimal Functional Approximation Using Fuzzy Rules"
DATE: Thursday, Oct. 29
TIME: 11 am
ROOM: FL246, Beckett Conference Room.
ABSTRACT:
It is known that fuzzy systems can approximate arbitrary functions,
that is, they are universal approximators. However, the proof does
not provide an algorithm to build a fuzzy system that approximates an
analytically defined function to an arbitrary precision with a minimum
number of fuzzy rules. The goal of this work is to find such an
algorithm.
The talk begins with a review of fuzzy sets and fuzzy rules, showing
how a given function may be approximated by a fuzzy system. I will
then describe a general approach that varies the membership function
of a fuzzy set to define an approximating function. The result is a
method that utilizes the information contained in the analytic
definition of a function, such as its first and second derivatives, to
build a fuzzy system that approximates it.
----------------------------------------------------------------------
Everyone is welcome to attend!