MULTILAYER FEEDFORWARD NEURAL NETWORKS WITH SINGLE POWERS-OF-2 WEIGHTS

A new algorithm for designing multilayer feedforward neural networks with single powers-of-two weights is presented in this correspondence. By applying this algorithm, the digital hardware implementation of such networks becomes easier as a result of the elimination of multipliers. This proposed algorithm consists of two stages. First, the network is trained by using the standard backpropagation algorithm. Weights are then quantized to single powers-of-two values, and weights and slopes of activation functions are adjusted adaptively to reduce sum of squared output errors to a specified level. Simulation results indicate that the multilayer feedforward neural networks with single powers-of-two weights obtained using the proposed algorithm have similar generalization performance as the original networks with continuous weights.