.png){kind=icon}
Return
Cache-assisted task offloading strategy based on multi-agent deep reinforcement learning
.png){kind=icon}
PRE
AI
DOI：10.1016/j.comnet.2025.111483.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Proposes a SAMADDPG-DCDOA method to jointly optimize task offloading and content caching in multi-user, multi-server MEC systems. • Models task offloading as an MDP and content caching as a knapsack problem. • Introduces a self-attention mechanism into MADDPG for improved state information prioritization. • Develops a dynamic cache decision algorithm to enhance caching efficiency and utilization. • Our proposed method reduces the long-term average user cost by up to 45.4% when compared to baseline algorithms.
Journal
IF:
4.6
Papers:
1.7K
Citations:
1.6W
Organization
No organization information available
Cited Papers
ShareSaveDistributed Edge Computing Offloading Algorithm Based on Deep Reinforcement Learning.png)
IEEE ACCESS
OAAI
IEEE ACCESSIF3.6
OAAIShareSaveShareSave
View more