The non-linear properties of Spike-Rate Perceptron with sub-Poisson input

We present more specific non-linear properties of Spike-Rate Perceptron with sub-Poisson inputs based on the diffusion approximation of renewal process, which employs both first and second statistical representation, i.e. the means, variances and correlations of the synaptic input. It shows that such perceptron is also able to perform various complex non-linear tasks, and to successfully solve the XOR problem. Here such perceptron offers the same advantage that includes both the mean and the variance of the input signal as Spike-Rate Perceptron.