Multiple-objective optimization resource scheduling algorithm for edge computing based on Deep Q-Network

To resolve the tension between edge computing service providers who aim to reduce energy use and users who prioritize enhanced service quality, we introduce an innovative edge computing resource allocation model utilizing a Deep Q-Network (DQN). This model dynamically reconciles energy usage with task completion time by modifying the emphasis placed on various optimization objectives. Experimental findings reveal that the proposed model successfully harmonizes these competing priorities, achieving substantial improvements over the benchmark algorithm in both energy efficiency and task execution time.