Neurosynaptic Computational Elements for Adaptive Transient Synchrony: Biophysical Accuracy versus Hardware Complexity

Abstract

In this paper, we examine electro-chemically accurate, multi-compartment, neurosynaptic computational elements, and analyze their complexity, accuracy, and flexibility in signal processing of a time-varying task. We evaluate distributed patterns of simultaneously firing neurons in space and time, and we establish a transient synchrony and homeostatic regulation mechanism upon the underlying synaptic connectivity. With synchronic spiking, we form synchronous groups of neuronal subpopulations, which represent content forming a coherent entity. The neurosynaptic computational elements implemented on Xilinx Virtex 7 XC7VX550 FPGA board illustrate feasibility of the methodology.