Neural Nets

  • James J. Buckley
  • Esfandiar Eslami
Part of the Advances in Soft Computing book series (AINSC, volume 13)


We first introduce layered, feedforward, neural nets in the next section and go through all the details on how they compute their output given inputs. Our applications of these neural nets is to: (1) approximate solutions (the α-cut and interval arithmetic solution of Section 5.2.3 in Chapter 5) to fuzzy equations, and (2) approximate the values (the α-cut and interval arithmetic value of Section 8.3 in Chapter 8) of fuzzy functions. In both cases the neural net requires sign constraints on its weights (some weights must be positive and the rest must be negative). Then in the third section of this chapter we fuzzify to get a fuzzy neural net. Our applications of fuzzy neural nets is to construct hybrid fuzzy neural nets for fuzzy functions. There is now no approximation, the output from the hybrid fuzzy neural nets will exactly equal the values of the fuzzy function.


Transfer Function Fuzzy Number Output Neuron Triangular Fuzzy Number Input Neuron 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • James J. Buckley
    • 1
  • Esfandiar Eslami
    • 2
  1. 1.Mathematics DepartmentUniversity of Alabama at BirminghamBirminghamUSA
  2. 2.Department of MathematicsShahid Bahonar UniversityKermanIran

Personalised recommendations