A New Algorithm Based on the Frank-Wolfe Method and Neural Network for a Class of Bilevel Decission Making Problems
一类二层决策问题的性质分析和基于Frank-Walfe方法的神经网络算法

In this paper a model for a class of bilevel decision-making problems is presented in the form of a static Stackelberg game. Under certain convexity assumptions, some properties of the model is studied. According to these properties, this paper presents a new algorithm based on the Frank-Wolfe method and neural network method. An example demonstrates the feasibility of the proposed algorithm.