Saved in:
Bibliographic Details
Main Authors: Torres, José Eduardo Zerna, Avgeris, Marios, Papagianni, Chrysa, Pongrácz, Gergely, Gódor, István, Grosso, Paola
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.13806
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916927164645376
author Torres, José Eduardo Zerna
Avgeris, Marios
Papagianni, Chrysa
Pongrácz, Gergely
Gódor, István
Grosso, Paola
author_facet Torres, José Eduardo Zerna
Avgeris, Marios
Papagianni, Chrysa
Pongrácz, Gergely
Gódor, István
Grosso, Paola
contents This work presents a proof-of-concept implementation of a distributed, in-network reinforcement learning (IN-RL) framework for adaptive path selection in programmable networks. By combining Stochastic Learning Automata (SLA) with real-time telemetry data collected via In-Band Network Telemetry (INT), the proposed system enables local, data-driven forwarding decisions that adapt dynamically to congestion conditions. The system is evaluated on a Mininet-based testbed using P4-programmable BMv2 switches, demonstrating how our SLA-based mechanism converges to effective path selections and adapts to shifting network conditions at line rate.
format Preprint
id arxiv_https___arxiv_org_abs_2508_13806
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Reinforcement Learning-based Adaptive Path Selection for Programmable Networks
Torres, José Eduardo Zerna
Avgeris, Marios
Papagianni, Chrysa
Pongrácz, Gergely
Gódor, István
Grosso, Paola
Machine Learning
This work presents a proof-of-concept implementation of a distributed, in-network reinforcement learning (IN-RL) framework for adaptive path selection in programmable networks. By combining Stochastic Learning Automata (SLA) with real-time telemetry data collected via In-Band Network Telemetry (INT), the proposed system enables local, data-driven forwarding decisions that adapt dynamically to congestion conditions. The system is evaluated on a Mininet-based testbed using P4-programmable BMv2 switches, demonstrating how our SLA-based mechanism converges to effective path selections and adapts to shifting network conditions at line rate.
title Reinforcement Learning-based Adaptive Path Selection for Programmable Networks
topic Machine Learning
url https://arxiv.org/abs/2508.13806