【摘要】 针对带等式约束的二次规划，提出了一类递归神经网络.首先，将二次规划的KKT条件转化成一个等价的方程组；然后对转化后的方程组，设计了一个递归神经网络.在理论上，利用Lyapunov稳定性理论，证明了该神经网络具有渐进收敛性.最后通过均值-方差投资组合选择问题验证了所设计神经网络的有效性.更多还原

【Abstract】 For quadratic programming with equality constraints, a class of recurrent neural network is proposed in this paper.Firstly, the KKT condition of quadratic programming is transformed into an equivalent system of equations; Then, a recurrent neural network is designed for the transformed equations.In theory, using Lyapunov stability theory, this paper proves that the neural network has asymptotic convergence.Finally, the effectiveness of the designed neural network is verified by mean variance portfolio selection.更多还原