An Efficient Solution Algorithm Based on Reinforcement Learning for Some NP Problems

Agent reinforcement learning is an important branch of machine learning.In this paper the author introduces the theory of Agent reinforcement learning algorithm,establishes a mathematic model to solve some NP problems such as Traveling Salesman Problem,and presents a framework and a basic method using Agent reinforcement learning algorithm to solve this class problem.We have solved a problem about tournament sorting with our novel algorithm successfully,and the result shows it is more effective than traditional methods.