Applications of Multi-Agent Deep Reinforcement Learning: Models and Algorithms

Recent advancements in deep reinforcement learning (DRL) have led to its application in multi-agent scenarios to solve complex real-world problems, such as network resource allocation and sharing, network routing, and traffic signal controls. Multi-agent DRL (MADRL) enables multiple agents to interact with each other and with their operating environment, and learn without the need for external critics (or teachers), thereby solving complex problems. Significant performance enhancements brought about by the use of MADRL have been reported in multi-agent domains; for instance, it has been shown to provide higher quality of service (QoS) in network resource allocation and sharing. This paper presents a survey of MADRL models that have been proposed for various kinds of multi-agent domains, in a taxonomic approach that highlights various aspects of MADRL models and applications, including objectives, characteristics, challenges, applications, and performance measures. Furthermore, we present open issues and future directions of MADRL.

Application of BP Neural Network to Optimize the Allocation of Art Teaching Resources

Reasonable allocation of art teaching resources can improve the management efficiency of art teaching resources. There is a large delay in the allocation of art teaching resources, which leads to the long occupation time of network resource allocation channel. The traditional method of network experiment resource allocation is to assign resource tasks for different channels to complete the resource allocation. When the network resource allocation channel occupies a long time, the allocation efficiency is reduced. This paper proposes an optimal allocation method of art teaching resources based on multi rate cognition. From the point of view that there are a pair of primary users and a pair of secondary users in the network, this method constructs a resource allocation delay model, obtains the resource allocation delay under different modes, and dynamically adjusts the transmission rate on the allocation resource block. The art teaching resource allocation scheduling problem is modeled as a nonlinear optimization problem, and the constraints of the optimization problem are given, which are integrated into greedy computing. The global optimal solution of the problem is carried out by using the method, and the allocation of art teaching resources is completed. Simulation results show that the proposed algorithm greatly improves the efficiency and effect of teaching network resource allocation.