Design of risk-sensitive optimal control for stochastic recurrent neural networks by using Hamilton-Jacobi-Bellman equation | IEEE Conference Publication | IEEE Xplore