Finite-time stabilization in probability for stochastic reaction–diffusion Cohen–Grossberg neural networks with mixed delays

We studied finite-time stabilization in probability for stochastic reaction-diffusion Cohen-Grossberg neural networks with mixed delays and multiplicative noise under Neumann boundary conditions. By developing a Lyapunov-Krasovskii functional and using stochastic theory with the Neumann-Poincaré inequality, we obtained several verifiable stabilization criteria. A componentwise controller with a nonlinear finite-time term was further designed to induce a constant drift effect in the Lyapunov evolution, leading to an explicit stochastic settling time bound that quantifies the roles of diffusion, delays, couplings, noise, and control gains. Simulations confirmed the effectiveness of the proposed method.