Global exponential stability for a class of generalized neural networks with distributed delays (Q2572144): Difference between revisions

The authors consider a generalized neural network model with distributed delay of the form \[ \begin{multlined} \dot{x}_i(t)=-d_i(x_i(t))+\sum\limits_{j=1}^nw_{ij}(x(t))f_j(x_j(t))\\ +\sum\limits_{j=1}^nw_{ij}^{\tau}(x(t))\int_{-\infty}^tK_{ij}(t-s)f_j(x_j(s))\,ds,\;\;i=1,\dots,n, \end{multlined} \] with \(x(t)=(x_1(t),\ldots,x_n(t))\), all functions are continuous, \(d_i(0)=0\) and \(f_j(0)=0\), \(f_j\) satisfies a global Lipschitz condition, \[ 0<\underline{D}_i\leq[d_i(u)-d_i(v)]/(u-v)\leq\overline{D}_i\text{ for }u\neq v, \] \[ K_{ij}(t)\geq0,\;\int_{0}^{\infty}K_{ij}(s)\,ds=1,\;\;\int_{0}^{\infty}sK_{ij}(s)\,ds<\infty,\quad \int_{0}^{\infty}K_{ij}(s)e^{\mu s}\,ds<\infty,\;\mu>0, \] \(i,j=1,\ldots,n\). Employing several Lyapunov functionals, the authors obtain various sufficient conditions on the global exponential stability of the zero solution. New results generalize and extend some previously known theorems.