基于深度强化学习的个性化任务处理方法

Personalized task processing method based on deep reinforcement learning

为实现移动云计算背景下多端个性化任务处理的目标,首先通过PCA等预处理方法萃取数据特征信息实现数据域分类。再通过深度强化学习方法结合多端模型的返回特征完成云计算中心个性化任务处理(PTP)方法模型的训练。实验及仿真结果表明,该方法不仅实现了个性化任务处理的目标,而且可以提高模型收敛速度,降低算法复杂度。

In order to achieve the goal of multiterminal personalized task processing in the context of mobile cloud computing,firstly,data domain classification is achieved by extracting data feature information through pre-processing methods such as PCA.Then complete the training of the personalized task processing(PTP)method model of cloud computing center through the deep reinforcement learning method combined with the return characteristics of the multi-terminal model.Experimental and simulation results show that this method not only achieves the goal of personalized task processing,but also improves the convergence speed of the model and reduces the complexity of the algorithm.