On the Functional Equivalence of TSK Fuzzy Systems to Neural Networks, Mixture of Experts, CART, and Stacking Ensemble Regression

Abstract

Fuzzy systems have achieved great success in numerous applications. However, there are still many challenges in designing an optimal fuzzy system, e.g., how to efficiently optimize its parameters, how to balance the trade-off between cooperations and competitions among the rules, how to overcome the curse of dimensionality, how to increase its generalization ability, etc. Literature has shown that by making appropriate connections between fuzzy systems and other machine learning approaches, good practices from other domains may be used to improve the fuzzy systems, and vice versa. This paper gives an overview on the functional equivalence between Takagi-Sugeno-Kang fuzzy systems and four classic machine learning approaches -neural networks, mixture of experts, classification and regression trees, and stacking ensemble regression -for regression problems. We also point out some promising new research directions, inspired by the functional equivalence, that could lead to solutions to the aforementioned problems. To our knowledge, this is so far the most comprehensive overview on the connections between fuzzy systems and other popular machine learning approaches, and hopefully will stimulate more hybridization between different machine learning algorithms.

Adaptation of fuzzy inference system to solve assessment problems of technical condition of construction objects

“…Moreover, the rapid development of hybrid technologies makes it possible to use fuzzy systems to display generalized expert knowledge on the architecture of artificial neural networks with their subsequent training on real data [11,12]. Thus, the use of neuro-fuzzy inference models makes it possible to automate the process of obtaining logical conclusions from input according to fuzzy rules specified by experts.…”

Hemodynamic Analysis for Olfactory Perceptual Degradation Assessment Using Generalized Type-2 Fuzzy Regression

“…OLFACTORY PERCEPTUAL-DEGRADATION This section provides detailed design of GT2FS based prediction for the assessment of subjective perceptualdegradation during the training and the test phases using Mamdami-like approach. The Mamdani-like formulation is required to utilize the consequent GT2FS MFs, which could not be used in case of Takagi-Sugeno-Kang (TSK) GT2FS model [45]. The Mamdani type GT2FS regression yields…”

Multiclass Fuzzily Weighted Adaptive-Boosting-Based Self-Organizing Fuzzy Inference Ensemble Systems for Classification

“…This ensemble aims to achieve high classification precision while maintaining its interpretability by utilising the so-called confidence-based voting strategy such that the fuzzy classifier serves as the main component and the second classifier will be activated only when the confidence level of the fuzzy classifier is low. In [46], the functional equivalence of Takagi-Sugeno-Kang (TSK) fuzzy systems to different regression approaches including stacking ensemble regression is studied, and it is shown that each IF-THEN rule in the fuzzy system is equivalent to a base model in stacking ensemble regression. Self-organising fuzzy inference system (SOFIS) [47], [48] is a recently introduced zero-order fuzzy inference system (FIS) for classification.…”