A computing unloading scheduling method based on depth reinforcement learning

Abstract

The invention provides a computing unloading scheduling method based on depth reinforcement learning, which provides a method for computing unloading to make unloading decision for Internet of Thingsequipment, including making decisions on various aspects needing to be unloaded according to the basic model of computing unloading. Based on different optimization objectives, different optimizationobjectives can be achieved by changing the value function. TheDeep-SARSA algorithm is similar to DQN algorithm, which combines reinforcement learning and depth learning. It can effectively change theunloading state and unloading action into training samples of depth learning when cooperating with experience pool. The invention can effectively carry out machine learning on an unloading state modelof an unlimited dimension, reducing the complexity of learning, this method uses neural network as the linear approximator of Q value, which can effectively improve the training speed and reduce thesample required for training. This method can effectively make the best decision through deep reinforcement learning under the given model and optimization objective.

Description

technical field The invention relates to a scheduling algorithm for computing offloading, in particular to a scheduling method for computing offloading based on deep reinforcement learning (Deep-SARSA), which belongs to the application of machine learning technology in the field of distributed computing. Background technique With the advancement of technology, applications pay more attention to the connection and interaction between the real world and the virtual world, such as face recognition, computer vision, natural language processing and other applications that require powerful computer capabilities. However, due to the inherent limitations of IoT devices (such as computing resources, batteries, memory, etc.), the tension between resource-constrained devices and such computing-intensive applications becomes a bottleneck in delivering satisfactory quality of experience (QoE), Therefore, the arrival of a mature mobile application market may be delayed. In this case, a n...

AI Technical Summary

Problems solved by technology

However, in some special and complex scenarios, because artificial intelligence cannot provide better features, only using RL will not be able to obtain better results, so it is necessary to introduce deep learning at this time.

Images