A REAL-TIME LEARNING ALGORITHM FOR A MULTILAYERED NEURAL NETWORK BASED ON THE EXTENDED KALMAN FILTER

The extended Kalman filter (EKF) is well known as a state estimation m ethod for a nonlinear system, and can be used as a parameter estimatio n method by augmenting the state with unknown parameters. A multilayer ed neural network is a nonlinear system having a layered structure, an d its learning algorithm is regarded as parameter estimation for such a nonlinear system. In this paper, a new real-time learning algorithm for a multilayered neural network is derived from the EKF. Since this EKF-based learning algorithm approximately gives the minimum variance estimate of the linkweights, the convergence performance is improved i n comparison with the backwards error propagation algorithm using the steepest descent techniques. Furthermore, tuning parameters which cruc ially govern the convergence properties are not included, which makes its application easier. Simulation results for the XOR and parity prob lems are provided.