CHARACTERISTICS OF HEBBIAN-TYPE ASSOCIATIVE MEMORIES WITH QUANTIZED INTERCONNECTIONS

The effects of quantized Hebbian-type associative memories (HAM's) on storage capacity and hardware implementations are explored in this pap er. The quantizations include a two-level strategy and a three-level s trategy using a cutoff threshold in which the interconnection weights of both strategies are restricted to binary memory points on each syna pse. The restricted interconnection weights enable a digital or an ana log HAM to be implemented more space efficiently. The three-level quan tized network with sparse connectivity among neurons has further simpl ified analog circuits in the dedicated hardwares. Results of simulatio n and theory show that the three-level quantized network with a proper ly selected cutoff threshold possesses higher storage capacity than th e two-level quantized network.