Complex-Valued GMDH-type Neural Network for Real-Valued Classification Problems

Recently, the application of complex-valued neural networks (CVNNs) for real-valued classification has attracted more and more attention. To overcome the limitations of the existing CVNNs, this study extends the real-valued group method of data handling (RGMDH) type neural network to complex domain, and constructs complex-valued GMDH-type neural network (CGMDH). First, it proposes the complex least squares for parameter estimation, and then constructs the complex external criterion to evaluate and select the middle candidate models. We conduct experiments in 10 UCI real-valued classification datasets. The results show that the performance of CGMDH is better than that of RGMDH and other four models. At the same time, the convergence speed of CGMDH is faster than that of RGMDH.