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 method for a nonlinear system, and can be used as a parameter estimation method by augmenting the state with unknown parameters. A multilayered neural network is a nonlinear system having a layered structure, and 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 in comparison with the backwards error propagation algorithm using the steepest descent techniques. Furthermore, tuning parameters which crucially govern the convergence properties are not included, which makes its application easier. Simulation results for the XOR and parity problems are provided.