Parallel, self-organizing hierarchical neural networks

A neural network architecture called the parallel self-organizing hierarchical neural network (PSHNN) is discussed. The PSHNN involves a number of stages in which each stage can be a particular neural network (SNN). At the end of each SNN, error detection is carried out, and a number of input vectors are rejected. Between two SNNs there is a nonlinear first SNN. The PSHNN has an optimized system complexity in the sense of minimized self-organizing number of stages, high classification accuracy, minimized learning and recall times, and parallel architectures in which all SNNs operate simultaneously without waiting for data from each other during testing. In classification experiments with aircraft and satellite remote-sensing data the PSHNN is compared to multilayer networks using backpropagation training.<
>