MAPPING OF ARTIFICIAL NEURAL NETWORKS ONTO MESSAGE-PASSING SYSTEMS

Various Artificial Neural Networks (ANN's) have been proposed in recen t years to mimic the human brain in solving problems involving human-l ike intelligence, Efficient mapping of ANN's comprising of large numbe r of neurons onto various distributed MIMD architectures is discussed in this paper, The massive interconnection among neurons demands a com munication-efficient architecture. Issues related to the suitability o f MIMD architectures for simulating neural networks are discussed, Per formance analysis of ring, torus, binary tree, hypercube, and extended hypercube for simulating artificial neural networks is presented. Our studies reveal that the performance of the extended hypercube is bett er than those of ring, torus, binary tree, and hypercube topologies.