Title :
Information Theoretic fuzzy modeling for regression
Author :
Álvarez-Estévez, Diego ; Príncipe, José C. ; Moret-Bonillo, Vicente
Author_Institution :
Lab. for the R&D of Artificial Intell. (LIDIA), Univ. of A Coruna, A Coruña, Spain
Abstract :
This paper presents a novel, Information Theoretic Learning (ITL) method to model a fuzzy system for regression tasks that minimizes the Renyi´s entropy of the error signal. An architecture based on a generalization of the well-known Adaptive-Network-Based Fuzzy Inference System (ANFIS) was used to perform such a modeling. The resulting method was tested on the prediction of future values for the Mackey-Glass chaotic time series. The results show that, when using the ITL cost function, the method returns better models in comparison with a Mean Squared Error (MSE)-guided cost function.
Keywords :
fuzzy systems; generalisation (artificial intelligence); inference mechanisms; mean square error methods; regression analysis; Mackey-Glass chaotic time series; Renyi entropy; adaptive-network-based fuzzy inference system; error signal; fuzzy system; generalization; information theoretic fuzzy modeling; information theoretic learning method; mean squared error; regression tasks; Clustering algorithms; Cost function; Entropy; Fuzzy systems; Input variables; Machine learning;
Conference_Titel :
Fuzzy Systems (FUZZ), 2010 IEEE International Conference on
Conference_Location :
Barcelona
Print_ISBN :
978-1-4244-6919-2
DOI :
10.1109/FUZZY.2010.5584499