Semi-Online Computational Offloading by Dueling Deep-Q Network for User Behavior Prediction
| Autor: | Hongyu Sun, Zhanyang Zhang, Chin-Ling Chen, Zhiyi Fang, Shinan Song |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
General Computer Science
Computer science Distributed computing General Engineering Edge computing dueling deep-Q network Computational resource user behavior prediction Dynamic simulation Task (computing) Resource (project management) Factor (programming language) Reinforcement learning computational offloading General Materials Science lcsh:Electrical engineering. Electronics. Nuclear engineering Activity-based costing computer lcsh:TK1-9971 computer.programming_language |
| Zdroj: | IEEE Access, Vol 8, Pp 118192-118204 (2020) |
| ISSN: | 2169-3536 |
| Popis: | Task offloading could optimize computational resource utilization in edge computing environments. However, how to assign and offload tasks for different behavior users is an essential problem since the systems dynamic, intelligent application diversity, and user personality. With user behavior prediction, this paper proposes soCoM, a semi-online Computational Offloading Model. We explore the user behaviors in sophisticated action space by reinforcement learning for catching unknown environment information. With Dueling Deep-Q Network, both the prediction accuracy of users' behaviors and the server load balance are well-considered, while increasing the computational efficiency and decreasing the resource costing. We propose a dynamic simulation environment of edge computing to demonstrate that user behavior is the critical factor for impacting system utilization. As the action space increasing, Dueling DQN performs better than state-of-art DQN and other improved strategies, and also load balance in multiple different server scenario. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|