Abstract:

While the Choquet integral (ChI) is a powerful parametric nonlinear aggregation function, it has limited scope and is not a universal function generator. Herein, we focus on a class of problems that are outside the scope of a single ChI. Namely, we are interested in tasks where different subsets of inputs require different ChIs. Herein, a genetic program (GP) is used to extend the ChI, referred to as GpChI hereafter, specifically in terms of compositions of ChIs and/or arithmetic combinations of ChIs. An algorithm is put forth to learn the different GP ChIs via genetic algorithm (GA) optimization. Synthetic experiments demonstrate GpChI in a controlled fashion, i.e., we know the answer and can compare what is learned to the truth. Real-world experiments are also provided for the multi-sensor fusion of electromagnetic induction (EMI) and ground penetrating radar (GPR) for explosive hazard detection. Our multi-sensor fusion experiments show that there is utility in changing aggregation strategy per different subsets of inputs (sensors or algorithms) and fusing those results.

Links:

Citation:

R. Smith, D. Anderson, A. Zare, J. Ball, B. Alvey, J. Fairley, and S. Howington, "Genetic Programming Based Choquet Integral for Multi-Source Fusion," in IEEE Int. Conf. Fuzzy Systems (FUZZ-IEEE), 2017.

@InProceedings{Smith2017Genetic,
Title = {Genetic Programming Based Choquet Integral for Multi-Source Fusion},
Author = {Ryan E. Smith and Derek T. Anderson and Alina Zare and John E. Ball and Brendan Alvey and Josh R Fairley and Stacy E Howington},
Booktitle = {IEEE Int. Conf. Fuzzy Systems (FUZZ-IEEE)},
Year = {2017},
Month = {Jul.}
}