“…Combine the output observation y and the full model candidates V 0,m j,k (t) , V r n ,m l,j,k (t) , the maximum likelihood estimatorα mle can be estimated by using a generalized linear model fitting method. The output spike probability g (θ (t)) can further be obtained using Equation (16). Assume G (t) =g (θ (t)) , t = 1, 2, · · · , T, to be a vector of the measured outputs with T time instants.…”