Dynamics of delayed cellular neural networks in the Stepanov pseudo almost automorphic space (Q2090495)

The paper deals with the model type cellular neural network with functional delays \[ x_i'(t)=-a_i(t)x_i(t)+\displaystyle\sum_{m=1}^n b_{ij}(t)f_j(x_j(t))+\displaystyle\sum_{m=1}^nc_{ij}(t)g_j(x_j(t-\xi_j(t)))+I_i(t),\tag{1} \] for \(t\in\mathbb{R}\) and \(i=1,2,\dots,n\), with the initial conditions \[ x_i(s)=v_i(s),\,\,\,s\in (-\infty,0],\,\,\,i=1,2,\dots,n,\tag{2} \] where \(i\) is the label of units in the neural networks model, \(x_i(t)\in\mathbb{R}\) represents the evolution state of the \(i\)th neuron at time \(t\), \(\xi_i(\cdot)\) is the signal delay at time \(t\) verifying \(t-\xi_i(t)\in\mathbb{R}\) for \(t\in \mathbb{R}\), \(I_i(\cdot)\) is the external bias of the \(i\)th unit, and \(v_i\in L_{\mathrm{loc}}^p((-\infty,0],\mathbb{R})\), for \(i=1,2,\dots,n\). The authors study the existence, uniqueness and the \(S^p\)-global exponential stability of Stepanov-like pseudo almost automorphy solution of problem \((1),(2)\).