Abstract :
No clear consensus exists about how to exploit the potential for massively parallel implementations of artificial neural networks. Three hardware implementations to demonstrate the key issues surrounding their use are: model specific integrated circuits for neural associative memories, self-organizing feature maps, and local cluster neural networks
Keywords :
ULSI; content-addressable storage; neural chips; neural net architecture; parallel architectures; self-organising feature maps; associative memories; local clustering; massively parallel processing; model specific integrated circuits; neural networks; self-organizing feature maps; ultra-large scale-integration architectures; Artificial neural networks; Associative memory; Biological system modeling; CMOS technology; Information processing; Integrated circuit modeling; Microelectronics; Neurons; Parallel processing; Ultra large scale integration;
