MLP-RBF: A Cooperative Multi-Modular Neural Network Application in High-Energy Physics

We present a cooperative multi-modular neural network architecture: a Multi-Layer Perceptron (MLP), followed by a Radial Basis Function network (RBF). We show that, in the LEP experiment of electron-positron collision run at CERN, this architecture was able to outperform both a simple multi-layer perceptron, a multi-modular MLP+LVQ (LVQ: Learning Vector Quantization) and MLP+RBF trained sequentially and a conventional technique (Discriminant Analysis).