Training method and system for spiking neural network

The invention provides a training method and system for a spiking neural network. The training method comprises the following steps: constructing an iterative leakage integral distribution neuron model; for the spiking neural network, calculating a pulse feature vector of a spiking neuron in a current network layer at the current moment, calculating the number of pulses in a feature channel according to the pulse feature vector at the current moment, adjusting the weight of the feature channel according to the number of the pulses, and updating the membrane potential of the iterative leakage integral distribution neuron model; according to the label of the training sample and the output of the iterative leakage integral issuing neuron model, constructing a loss function used for training the spiking neural network; training is carried out in a back propagation mode, and the derivative of back propagation is calculated according to the membrane potential updated in the step 2. 本发明提出了一种用于脉冲神经网络的训练方法和系统，包括构建迭代泄露