EFFICIENT MAPPING OF BACKPROPAGATION ALGORITHM ONTO A NETWORK OF WORKSTATIONS

In this paper, we present an efficient technique for mapping a backpro pagation (BP) learning algorithm for multilayered neural networks onto a network of workstations (NOW's). We adopt a vertical partitioning s cheme, where each layer in the neural network is divided into p disjoi nt partitions, and map each partition onto an independent workstation in a network of p workstations. We present a fully distributed version of the BP algorithm and also its speedup analysis. We compare the per formance of our algorithm with a recent work involving the vertical pa rtitioning approach for mapping the BP algorithm onto a distributed me mory multiprocessor. Our results on SUN 3/50 NOW's show that we are ab le to achieve better speedups by using only two communication sets and also by avoiding some redundancy in the weights computation for one t raining cycle of the algorithm.