Graph transformers and stabilized reinforcement learning for large-scale dynamic routing modulation and spectrum allocation in elastic optical networks

Reinforcement learning (RL) has been widely applied to dynamic routing, modulation, and spectrum assignment (RMSA) in optical networks, yet no prior work has trained a transformer model for this task. We attribute this to the high data and compute requirements of transformers and potential training instabilities with RL. We address this gap by combining recent advances from the machine learning literature (rotary positional encodings for graph-structured data, off-policy invalid action masking, and valid mass regularization) with GPU-accelerated simulation to achieve, for the first time to our knowledge, stable RL training of a transformer for dynamic RMSA. We demonstrate, through systematic benchmarking against previous RL methods and heuristic algorithms, that ours is the first RL method to exceed all benchmarks, increasing the supportable traffic load by up to 13%. To demonstrate the scalability of our approach, we train on real network topologies from the TopologyBench database up to 143 nodes and 362 links, with 320×12.5GHz frequency slot units per link, and 100 Gbps traffic requests. To our knowledge, these are the largest dynamic RMSA problems to which RL has been applied. We find that up to 4% increased traffic load can be supported at low blocking probability (<0.1%) with our method compared to the best available benchmark algorithm. We present an ablation study of the components of our training algorithm, the dynamics of the loss function during training, and an analysis of the allocation decisions of the trained models. We make all code used to produce this paper openly available for reproduction and future benchmarking.

Authors

Institutions

Publication Details

Journal
Journal of Optical Communications and Networking
Published
2026-11-17
DOI
https://doi.org/10.1364/jocn.605888
Primary Topic
Advanced Optical Network Technologies
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Graph transformers and stabilized reinforcement learning for large-scale dynamic routing modulation and spectrum allocation in elastic optical networks

Alejandra Beghelli, Laura Toni, Michael Doherty
4 links
Journal of Optical Communications and Networking
Advanced Optical Network Technologies
article

Graph transformers and stabilized reinforcement learning for large-scale dynamic routing modulation and spectrum allocation in elastic optical networks

Alejandra Beghelli, Laura Toni, Michael Doherty
article en

Abstract

Reinforcement learning (RL) has been widely applied to dynamic routing, modulation, and spectrum assignment (RMSA) in optical networks, yet no prior work has trained a transformer model for this task. We attribute this to the high data and compute requirements of transformers and potential training instabilities with RL. We address this gap by combining recent advances from the machine learning literature (rotary positional encodings for graph-structured data, off-policy invalid action masking, and valid mass regularization) with GPU-accelerated simulation to achieve, for the first time to our knowledge, stable RL training of a transformer for dynamic RMSA. We demonstrate, through systematic benchmarking against previous RL methods and heuristic algorithms, that ours is the first RL method to exceed all benchmarks, increasing the supportable traffic load by up to 13%. To demonstrate the scalability of our approach, we train on real network topologies from the TopologyBench database up to 143 nodes and 362 links, with 320×12.5GHz frequency slot units per link, and 100 Gbps traffic requests. To our knowledge, these are the largest dynamic RMSA problems to which RL has been applied. We find that up to 4% increased traffic load can be supported at low blocking probability (<0.1%) with our method compared to the best available benchmark algorithm. We present an ablation study of the components of our training algorithm, the dynamics of the loss function during training, and an analysis of the allocation decisions of the trained models. We make all code used to produce this paper openly available for reproduction and future benchmarking.

Journal of Optical Communications and NetworkingVol. 18(12)
BT Group (United Kingdom) (GB)
Engineering and Physical Sciences Research Council
Industry, innovation and infrastructure
Advanced Optical Network Technologies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon