FAuNO: Semi-Asynchronous Federated Reinforcement Learning Framework for Task Offloading in Edge Systems

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Metelo, Frederico, Oliveira, Alexandre, Racković, Stevo, Costa, Pedro Ákos, Soares, Cláudia
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912933257150464
author Metelo, Frederico
Oliveira, Alexandre
Racković, Stevo
Costa, Pedro Ákos
Soares, Cláudia
author_facet Metelo, Frederico
Oliveira, Alexandre
Racković, Stevo
Costa, Pedro Ákos
Soares, Cláudia
contents Edge computing addresses the growing data demands of connected-device networks by placing computational resources closer to end users through decentralized infrastructures. This decentralization challenges traditional, fully centralized orchestration, which suffers from latency and resource bottlenecks. We present \textbf{FAuNO} -- \emph{Federated Asynchronous Network Orchestrator} -- a buffered, asynchronous \emph{federated reinforcement-learning} (FRL) framework for decentralized task offloading in edge systems. FAuNO adopts an actor-critic architecture in which local actors learn node-specific dynamics and peer interactions, while a federated critic aggregates experience across agents to encourage efficient cooperation and improve overall system performance. Experiments in the \emph{PeersimGym} environment show that FAuNO consistently matches or exceeds heuristic and federated multi-agent RL baselines in reducing task loss and latency, underscoring its adaptability to dynamic edge-computing scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2506_02668
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FAuNO: Semi-Asynchronous Federated Reinforcement Learning Framework for Task Offloading in Edge Systems
Metelo, Frederico
Oliveira, Alexandre
Racković, Stevo
Costa, Pedro Ákos
Soares, Cláudia
Artificial Intelligence
Machine Learning
Edge computing addresses the growing data demands of connected-device networks by placing computational resources closer to end users through decentralized infrastructures. This decentralization challenges traditional, fully centralized orchestration, which suffers from latency and resource bottlenecks. We present \textbf{FAuNO} -- \emph{Federated Asynchronous Network Orchestrator} -- a buffered, asynchronous \emph{federated reinforcement-learning} (FRL) framework for decentralized task offloading in edge systems. FAuNO adopts an actor-critic architecture in which local actors learn node-specific dynamics and peer interactions, while a federated critic aggregates experience across agents to encourage efficient cooperation and improve overall system performance. Experiments in the \emph{PeersimGym} environment show that FAuNO consistently matches or exceeds heuristic and federated multi-agent RL baselines in reducing task loss and latency, underscoring its adaptability to dynamic edge-computing scenarios.
title FAuNO: Semi-Asynchronous Federated Reinforcement Learning Framework for Task Offloading in Edge Systems
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2506.02668