# Adaptive AI-based Decentralized Resource Management in the Cloud-Edge Continuum

arXiv: https://arxiv.org/abs/2501.15802 · HTML: https://mlsanity.com/paper/2501.15802/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 33 references that are in the map.
- First posted on arXiv 27 Jan 2025.

## Builds on (most cited first, 9 of 9)

- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [Resource Management in Wireless Networks via Multi-Agent Deep Reinforcement Learning](https://mlsanity.com/paper/2002.06215/) (2020)
- [Graph Neural Networks and Deep Reinforcement Learning Based Resource Allocation for V2X Communications](https://mlsanity.com/paper/2407.06518/) (2024)
- [Resource Management in Fog/Edge Computing: A Survey](https://mlsanity.com/paper/1810.00305/) (2018)
- [YAFS: A simulator for IoT scenarios in fog computing](https://mlsanity.com/paper/1902.01091/) (2019)
- [SPQR: Controlling Q-ensemble Independence with Spiked Random Model for Reinforcement Learning](https://mlsanity.com/paper/2401.03137/) (2024)
- [The Internet of Things, Fog and Cloud Continuum: Integration and Challenges](https://mlsanity.com/paper/1809.09972/) (2018)
- [EdgeSlice: Slicing Wireless Edge Computing Network with Decentralized Deep Reinforcement Learning](https://mlsanity.com/paper/2003.12911/) (2020)
- [Socio-technical Smart Grid Optimization via Decentralized Charge Control of Electric Vehicles](https://mlsanity.com/paper/1701.06811/) (2017)

## Built on by (most cited first, 1 of 1)

- [A Hybrid Proactive And Predictive Framework For Edge Cloud Resource Management](https://mlsanity.com/paper/2511.16075/) (2025)

## Abstract (arXiv)

In the Cloud-Edge Continuum, dynamic infrastructure change and variable workloads complicate efficient resource management. Centralized methods can struggle to adapt, whilst purely decentralized policies lack global oversight. This paper proposes a hybrid framework using Graph Neural Network (GNN) embeddings and collaborative multi-agent reinforcement learning (MARL). Local agents handle neighbourhood-level decisions, and a global orchestrator coordinates system-wide. This work contributes to decentralized application placement strategies with centralized oversight, GNN integration and collaborative MARL for efficient, adaptive and scalable resource management.