An application-specific array architecture for feedforward with backpropagation ANNs
Qutaibah Marwan Malluhi, Magdy Bayoumi, T.R.N. Rao · 2002
An application-specific array architecture for Artificial Neural Networks (ANNs) computation is proposed. This array is configured as a mesh-of-appendixed-trees (MAT). Algorithms to implement both the recall and the training phases of the multilayer feedforward with backpropagation ANN model are developed on MAT. The proposed MAT architecture requires only O(log N) time, while other reported techniques offer O(N) time, where N is the size of the largest layer. Beside the high speed performance, pipelining of more than one input pattern can be achieved which further improves the performance.>