A logarithmic-approximation approach for bandwidth efficient MoE deployment in edge networks
Authors:
- Xiangning LuXLXiangning Lu
- Xiaohui LiuXLXiaohui Liu
- Chao WangCWChao Wang
- Jiaqi RenJRJiaqi Ren
- Danyang ZhengDZDanyang Zheng
- Xiangyi ChenXCXiangyi Chen
- Huanlai XingHXHuanlai Xing
- 21 July 2026
- journal article
- Published by Elsevier in Computer Networks
- Vol. 287
Abstract
No abstract availableFunding Information
- National Natural Science Foundation of China (62302404)
- Fundamental Research Funds for the Central Universities (2682025GH021)
- Fundamental Research Funds for the Central Universities (2682026ZT007)
This publication has 20 references indexed in Scilit:
- A Runtime- and Cost-Efficient Approach of Deploying Mixture-of-Experts in Edge Networks2025
- IA-MOE: An Intensity-Aware Mixture of Experts for DNN Inference2025
- The MoE-Empowered Edge LLMS Deployment: Architecture, Challenges, and OpportunitiesIEEE Communications Magazine, 2025
- Towards Expert Models Deployment Cost Optimization in Edge Computing Networks2025
- Digital Twin Empowered Task Offloading for Mobile-Edge Computing in 6G Internet of VehiclesIEEE Internet of Things Journal, 2025
- Intelligent Task Offloading and Resource Allocation in Knowledge Defined Edge Computing NetworksIEEE Transactions on Mobile Computing, 2024
- Deep Reinforcement Learning for Dependent Task Offloading in Multi-Access Edge ComputingIEEE Access, 2024
- Task offloading strategies for mobile edge computing: A surveyComputer Networks, 2024
- Security-Aware Service Function Chaining and Embedding With Asymmetric Dedicated ProtectionIEEE Access, 2024
- Virtual Network Embedding with Virtual Nodes Ranking and Multi Points Sampling2022