Multi-strategy Collaborative Decision-making System Based On Deep Reinforcement Learning

In recent years,with the advent of a new wave of artificial intelligence,the research on AI has entered a brand new stage.Deep learning technologies,such as deep neural networks,are increasingly used in research and production.The artificial intelligence decision-making system based on deep reinforcement learning has huge application prospects in the fields of medical treatment and industrial production.Nevertheless,there are still some problems in the current research,such as the low training efficiency of deep learning models and the ignorant of the impact of environmental changes on model performance.In response to these problems,this paper proposes the idea of multi-strategy collaborative decision-making based on deep reinforcement learning,hoping to improve the training efficiency and performance stability of the decision-making model in a changing environment.The main contributions of this paper are as follows:1.Combined with the current research results,we explain the basic principles and research progress of deep reinforcement learning.Then we analyze the advantages and disadvantages of common deep reinforcement learning algorithms.The difficulties in the research of artificial intelligence decision-making systems are finally pointed out.2.Aimed at the problems of the current decision-making system,such as low training efficiency and insufficient consideration of the impact of changes in the decision environment,we design a machine decision model and an artificial decision model.Based on the idea of integrated learning,we propose a Multi-strategy collaborative decision algorithm combining artificial decision model and machine decision model.3.Based on the idea of multi-strategy collaborative decision-making,a variety of deep learning models and artificial decision-making models are fused to improve the performance of the model.The experimental results show that the multi-strategy collaborative decision-making model using convolutional neural network and residual network is significantly better than that only using convolutional neural networks.4.An artificial intelligence decision-making system is designed.Through this system,the various influencing factors in the artificial decision-making algorithm are tested and ranked,and the theoretical basis of the situation evaluation formula is given.On this basis,we compares the performance of the multi-strategy collaborative decision-making algorithm and the traditional deep reinforcement learning algorithm.The experimental results show that the proposed multi-strategy cooperative decision-making algorithm is superior to neural network-based deep reinforcement learning algorithms in four aspects: training speed,level of playing chess,mean square error,and model's adaptability to environmental changes.