Rapid feedforward computation by temporal encoding and learning with spiking neurons

As we know, primates perform remarkably well in cognitive tasks such as pattern recognition. Motivated from recent findings in biological systems, a unified and consistent feedforward system network with a proper encoding scheme and supervised temporal rules is built for processing real-world stimuli. The temporal rules are used for processing the spatiotemporal patterns. To utilize these rules on images or sounds, a proper encoding method and a unified computational model with consistent and efficient learning rule are required. Through encoding, external stimuli are converted into sparse representations which also have properties of invariance. These temporal patterns are then learned through biologically derived algorithms in the learning layer, followed by the final decision presented through the readout layer. The performance of the model is also analyzed and discussed. This chapter presents a general structure of SNN for pattern recognition, showing that the SNN has the ability to learn the real-world stimuli.