A bidirectional associative memory based on cortical spiking neurons using temporal coding

In this paper, we present a spiking bidirectional associative memory (BAM) with temporal coding. The coding scheme used in artificial neural networks (ANN) known as “mean firing rate” cannot comply with the fast and complex computations occurring in the cortex. In biological neural networks the information is coded and processed based on the timing of action potentials. To improve the biological plausibility of the standard BAM, we employ spiking neurons for its processing units, and information is presented to the model in the form of temporal coding. The neurons used in the model are heterogeneous, and capable to generate various spike-timing patterns. Genetic Algorithm and Co-evolution are used for training, and the experimental results of the proposed BAM are compared to those of the standard BAM. The results show improvements in recall, storage capacity and convergence which are of interest to design a BAM.