Multi-Task Learning-Based Task Scheduling Switcher for a Resource-Constrained IoT System

In this journal, we proposed a novel method of using multi-task learning to switch the scheduling algorithm. With change algorithm inside framework, framework can create scheduler with best task execution optimization under computation deadline. changing number tasks, types resources taken, and deadline, it is hard for single achieve while avoiding worst-case time complexity in resource-constrained Internet Things (IoT) system due trade-off each Furthermore, different hardware specifications affect differently, making rely on Big-O as reference. profile behavior used compute scheduler, identify Our benchmark result shows that an average 93.68% accuracy meeting along 23.41% optimization. Based results, our improve IoT system.