Brain-inspired pattern classification with memristive neural network using the Hodgkin-Huxley neuron

Abstract

Recent findings about using memristor devices to mimic biological synapses in neuromorphic systems open a new vision in neuroscience. Ultra-dense learning architectures can be implemented through the Spike-Timing-Dependent-Plasticity (STDP) mechanism by exploiting these nanoscale nonvolatile devices. In this paper, a Spiking Neural Network (SNN) that uses biologically plausible mechanisms is implemented. The proposed SNN relies on Hodgkin-Huxley neurons and memristor-based synapses to implement a bio-inspired neuromorphic platform. The behavior of the proposed SNN and its learning mechanism are discussed, and test results are provided to show the effectiveness of the proposed design for pattern classification applications.