Deriving multistage FNN models from Takagi and Sugeno's fuzzy systems

Fu Lai Korris Chung, Ji cheng Duan, Daniel So Yeung

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

5 Citations (Scopus)

Abstract

Two multistage fuzzy neural network (FNN) models are derived from Takagi and Sugeno's fuzzy systems by arranging single-stage reasoning units (stages) in an incremental and aggregation manner. The dimensionality problem is overcome since the number of rules is reduced to a linear function of the number of inputs. The network structure in each stage is based on Jang's ANFIS model. Through applying LSE and back-propagation algorithms to the training process, the proposed models can learn multistage fuzzy rules from stipulated data pairs. Simulation results show that the proposed multistage FNN models are superior to its single-stage counterpart in resource used (e.g. number of fuzzy rules, number of adjustable parameters etc.), convergence speed and generalization ability.
Original languageEnglish
Title of host publicationIEEE International Conference on Fuzzy Systems
PublisherIEEE
Pages1259-1263
Number of pages5
Publication statusPublished - 1 Jan 1998
EventProceedings of the 1998 IEEE International Conference on Fuzzy Systems,. Part 2 (of 2) - Anchorage, AK, United States
Duration: 4 May 19989 May 1998

Conference

ConferenceProceedings of the 1998 IEEE International Conference on Fuzzy Systems,. Part 2 (of 2)
Country/TerritoryUnited States
CityAnchorage, AK
Period4/05/989/05/98

ASJC Scopus subject areas

  • Chemical Health and Safety
  • Software
  • Safety, Risk, Reliability and Quality

Cite this