AutoTSK: Automatic Rule Enhancement via Sparse Gate Function in TSK Fuzzy Systems for High-Dimensional Classification

This repository provides the official Python implementation of the paper
"AutoTSK: Automatic Rule Enhancement via Sparse Gate Function in TSK Fuzzy Systems for High-Dimensional Classification".
AutoTSK introduces a sparse gate mechanism to automatically enhance rule selection in Takagi-Sugeno-Kang (TSK) fuzzy systems, aiming to improve performance in high-dimensional classification tasks.

📄 Paper

For a detailed explanation of the methodology and experimental results, please refer to our paper:
AutoTSK: Automatic Rule Enhancement via Sparse Gate Function in TSK Fuzzy Systems for High-Dimensional Classification

🙏 Acknowledgments

This project builds upon the functionalities provided by the open-source library PyTSK. PyTSK offers a Python toolbox for developing TSK fuzzy systems, supporting both fuzzy clustering and mini-batch gradient descent (MBGD) optimization techniques. Its integration with scikit-learn and PyTorch facilitates the construction and training of TSK models, especially for high-dimensional data scenarios.