METHOD AND APPARATUS FOR TRAINING MULTI-LAYER SPIKING NEURAL NETWORK

摘要

The disclosed multi-layer spiking neural network learning method generates, for each of the hidden layers of the multi-layer spiking neural network, input spikes of current hidden layer neurons connected to the previous hidden layer neurons based on the output spike frequency calculated in the previous hidden layer neurons. And, after STDP (Spike Timing Dependent Plasticity) learning of the current hidden layer neurons receiving the input spikes, the process of storing the output spike frequency calculated in the current hidden layer neurons for the input spikes is repeated. In the multilayer spiking neural network learning method, in the multilayer spiking neural network in which the STDP learning for the hidden layers is completed, the sum of the minimum synaptic weights related to firing for each neuron constituting the hidden layers is calculated, and the minimum synaptic weight sum for each neuron Based on this, the membrane potential threshold voltage of the corresponding neuron is adjusted.