On-line stabilization of block-diagonal recurrent neural networks

This paper deals with a discrete-time recurrent neural network (DTRNN) with a block-diagonal feedback weight matrix, called the block-diagonla recurre nt neural network (BDRNN), that allows a simplified approach to online trai ning and to address network and training stability issues. The structure of the BDRNN is exploited to modify the conventional backpropagation through time (BPTT) algorithm. to reduce its storage requirement by a numerically s table method of recomputing the network state variables, The network and tr aining stability is addressed by exploiting the BDRNN structure to directly monitor and maintain stability during weight updates by developing a funct ional measure of system stability that augments the cost function being min imized, Simulation results are presented to demonstrate the performance of the BDRNN architecture, its training algorithm, and the stabilization metho d.