EUVIMEDEUVIMED entdecken
Crossref

Hierarchical Meta-Graph Reinforcement Learning for Collaborative GenAI Model Caching and Inference Scheduling

Liang Zhao, Jing Wei, Huan Zhou, Zhen Chen, Tong Wu, Victor C. M. Leung

IEEE Transactions on Cognitive Communications and Networking · 2026 · 12 · 10216-10231

DOI: 10.1109/tccn.2026.3713352 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Cognitive Communications and Networking, 2026. Verfasst von Liang Zhao und 5 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 43 Literaturangaben. Gefördert von: Hubei Provincial Natural Science Foundation of China, Open Fund of Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Enabling collaborative generative AI (GenAI) inference at the network edge is challenging due to limited caching capacity, heterogeneous computing resources, and highly dynamic, latency-sensitive service demands. In this paper, we investigate the joint optimization of GenAI model caching, inference offloading, and resource allocation in a collaborative cloud–edge–end architecture. To address the strong coupling between long-term caching decisions and short-term scheduling dynamics, we propose a Hierarchical Meta-Graph Reinforcement Learning framework, termed HMGRL. Specifically, a heat-greedy model caching strategy is developed to capture time-varying model popularity and to reduce switching overhead on a slow timescale, while a graph-enhanced dueling deep reinforcement learning algorithm with prioritized experience replay enables topology-aware collaborative inference offloading and resource allocation on a fast timescale. Extensive simulations demonstrate that HMGRL consistently outperforms representative baselines in terms of system utility, cache and computing-resource utilization, convergence stability, and performance robustness. These results validate the effectiveness of the proposed hierarchical learning framework for practical GenAI applications at the network edge.

Förderung

  • Hubei Provincial Natural Science Foundation of China
  • Open Fund of Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering
  • National Natural Science Foundation of China
  • Natural Science Foundation of Guangdong Province

Zitieren

APA

Zhao, L., Wei, J., Zhou, H., Chen, Z., Wu, T., & Leung, V. C. M. (2026). Hierarchical Meta-Graph Reinforcement Learning for Collaborative GenAI Model Caching and Inference Scheduling. IEEE Transactions on Cognitive Communications and Networking, 12, 10216-10231. https://doi.org/10.1109/tccn.2026.3713352

Vancouver

Zhao L, Wei J, Zhou H, Chen Z, Wu T, Leung VCM. Hierarchical Meta-Graph Reinforcement Learning for Collaborative GenAI Model Caching and Inference Scheduling. IEEE Transactions on Cognitive Communications and Networking. 2026;12:10216-10231. doi:10.1109/tccn.2026.3713352

BibTeX
@article{Zhao2026,
  author = {Zhao, Liang and Wei, Jing and Zhou, Huan and Chen, Zhen and Wu, Tong and Leung, Victor C. M.},
  title = {Hierarchical Meta-Graph Reinforcement Learning for Collaborative GenAI Model Caching and Inference Scheduling},
  journal = {IEEE Transactions on Cognitive Communications and Networking},
  year = {2026},
  volume = {12},
  pages = {10216--10231},
  doi = {10.1109/tccn.2026.3713352}
}
Kostenlos starten →