Zadeh's Type-2 Fuzzy Logic Systems: Precision and High-Quality Prediction Intervals

TL;DR

This paper revisits Zadeh's Type-2 Fuzzy Logic Systems, introducing a new design that enhances flexibility and accuracy in uncertainty quantification, especially for high-risk decision-making tasks, by integrating deep learning and addressing high-dimensional data challenges.

Contribution

It proposes a novel Zadeh's GT2 Fuzzy Set-based framework that improves design flexibility and learning efficiency in high-dimensional data, enabling high-quality prediction intervals.

Findings

Z-GT2-FLS achieves higher precision and reliable prediction intervals.

The framework effectively handles high-dimensional data and integrates deep learning optimizers.

Z-GT2-FLS outperforms existing GT2 and IT2 fuzzy systems in uncertainty quantification.

Abstract

General Type-2 (GT2) Fuzzy Logic Systems (FLSs) are perfect candidates to quantify uncertainty, which is crucial for informed decisions in high-risk tasks, as they are powerful tools in representing uncertainty. In this paper, we travel back in time to provide a new look at GT2-FLSs by adopting Zadeh's (Z) GT2 Fuzzy Set (FS) definition, intending to learn GT2-FLSs that are capable of achieving reliable High-Quality Prediction Intervals (HQ-PI) alongside precision. By integrating Z-GT2-FS with the \(\alpha\)-plane representation, we show that the design flexibility of GT2-FLS is increased as it takes away the dependency of the secondary membership function from the primary membership function. After detailing the construction of Z-GT2-FLSs, we provide solutions to challenges while learning from high-dimensional data: the curse of dimensionality, and integrating Deep Learning (DL) optimizers. We develop a DL framework for learning dual-focused Z-GT2-FLSs with high performances. Our study includes statistical analyses, highlighting that the Z-GT2-FLS not only exhibits high-precision performance but also produces HQ-PIs in comparison to its GT2 and IT2 fuzzy counterparts which have more learnable parameters. The results show that the Z-GT2-FLS has a huge potential in uncertainty quantification.