Large-scale neuromorphic computing with application in intelligent text recognition on many-core cluster

The development of a computer as something more than a calculating machine marked the birth of the field of pattern recognition. There has been increased interest in research involving use of machines for performing intelligent tasks normally associated with human behavior. This work is confined to a study of artificial neural systems from a computational intelligence perspective. To accomplish computational intensive artificial neural network system we adopt current technology i.e. Cell Broadband Engine Processor (Cell BE) which potentially allows us to build systems of a complexity that approaches the number of elements and interconnections of the brain. The Cell BE is used to accelerate the Brain-State-in-a-Box (BSB) recall operation. The BSB model, an associative neural network was first described by Anderson et al. (1977) is used primarily to model the effects of human cognizance functions. This work shows the implementation details of 256-neuron BSB model on Cell BE. The pattern recognition application was extensively tested on 256-neuron BSB model which gave promising results for noisy patterns. Experimental results show that, implementation on Cell BE is faster than the Intel Pentium IV.