Searcharxiv⌕ Search

arXiv subjects

Alejandra Beghelli

Publications and source records attributed to Alejandra Beghelli.

9 recordsLinked to original sources
arXiv2026

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, 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 x 12.5 GHz 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 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 analyze the allocation decisions of the trained models. We make all code used to produce this paper openly available for reproduction and future benchmarking: https://github.com/micdoh/XLRON.

cs.NI↗
arXiv2026

Impact of Topology on Multipartite Entanglement Distribution Protocols in Quantum Networks

Quantum networks will rely on entanglement distribution to enable multi-user applications such as distributed quantum computing and cryptography. While multipartite entanglement distribution routing protocols have been extensively studied on idealised grid topologies, less is understood about how real network structure shapes their performance and resource requirements. We present a systematic study of four routing protocols for multipartite entanglement distribution, each characterised by the number of paths (single-path and multi-path) and routing strategy (star-based and tree-based), over 81 real network topologies. We identified four distinct topology-dependent performance regimes, where: (i) all protocols perform poorly, (ii) tree-based protocols dominate, (iii) multi-path protocols dominate, or (iv) all protocols perform well. By correlating clusters with graph metrics, we also provide structural explanations for the varied performance of specific protocols. Additionally, motivated by the anticipated high cost of repeaters, we investigated the impact of repeater trimming on the performance of multi-path protocols. Topology strongly governs how far repeater nodes can be removed from the network while maintaining a given performance (distribution rate). For instance, in networks where only 80% of nodes operate as repeaters, well-performing topologies are able to retain over 90% of the distribution rate; whereas sparse, weakly connected graphs exhibit rapid performance degradation, retaining less than half of the distribution rate. Our results provide a topology-aware framework for protocol selection and infrastructure optimisation in future quantum networks, bridging routing design with cost-aware deployment strategies.

quant-ph↗
arXiv2025

Fidelity-Aware Multipath Routing for Multipartite State Distribution in Quantum Networks

We consider the problem of distributing entangled multipartite states across a quantum network with improved distribution rate and fidelity. For this, we propose fidelity-aware multi-path routing protocols, assess their performance in terms of the rate and fidelity of the distributed Greenberger-Horne-Zeilinger (GHZ) states, and compare such performance against that of single-path routing. Simulation results show that the proposed multi-path routing protocols select routes that require more Bell states compared to single-path routing, but also require fewer rounds of Bell state generation. We also optimised the trade-off between distribution rate and fidelity by selecting an appropriate cutoff to the quantum memory storage time. Using such a cutoff technique, the proposed multi-path protocols can achieve up to an 8.3 times higher distribution rate and up to a 28% improvement in GHZ state fidelity compared to single-path routing. These results show that multi-path routing both improves the distribution rates and enhances fidelity for multipartite state distribution.

quant-ph↗
arXiv2025

Reinforcement Learning with Graph Attention for Routing and Wavelength Assignment with Lightpath Reuse

Many works have investigated reinforcement learning (RL) for routing and spectrum assignment on flex-grid networks but only one work to date has examined RL for fixed-grid with flex-rate transponders, despite production systems using this paradigm. Flex-rate transponders allow existing lightpaths to accommodate new services, a task we term routing and wavelength assignment with lightpath reuse (RWA-LR). We re-examine this problem and present a thorough benchmarking of heuristic algorithms for RWA-LR, which are shown to have 6% increased throughput when candidate paths are ordered by number of hops, rather than total length. We train an RL agent for RWA-LR with graph attention networks for the policy and value functions to exploit the graph-structured data. We provide details of our methodology and open source all of our code for reproduction. We outperform the previous state-of-the-art RL approach by 2.5% (17.4 Tbps mean additional throughput) and the best heuristic by 1.2% (8.5 Tbps mean additional throughput). This marginal gain highlights the difficulty in learning effective RL policies on long horizon resource allocation tasks.

cs.NI↗
arXiv2025

Reinforcement Learning for Dynamic Resource Allocation in Optical Networks: Hype or Hope?

The application of reinforcement learning (RL) to dynamic resource allocation in optical networks has been the focus of intense research activity in recent years, with almost 100 peer-reviewed papers. We present a review of progress in the field, and identify significant gaps in benchmarking practices and reproducibility. To determine the strongest benchmark algorithms, we systematically evaluate several heuristics across diverse network topologies. We find that path count and sort criteria for path selection significantly affect the benchmark performance. We meticulously recreate the problems from five landmark papers and apply the improved benchmarks. Our comparisons demonstrate that simple heuristics consistently match or outperform the published RL solutions, often with an order of magnitude lower blocking probability. Furthermore, we present empirical lower bounds on network blocking using a novel defragmentation-based method, revealing that potential improvements over the benchmark heuristics are limited to 19-36% increased traffic load for the same blocking performance in our examples. We make our simulation framework and results publicly available to promote reproducible research and standardized evaluation https://doi.org/10.5281/zenodo.12594495.

cs.NI↗
arXiv2024

Topology Bench: Systematic Graph Based Benchmarking for Core Optical Networks

Topology Bench is a comprehensive topology dataset designed to accelerate benchmarking studies in optical networks. The dataset, focusing on core optical networks, comprises publicly accessible and ready-to-use topologies, including (a) 105 georeferenced real-world optical networks and (b) 270,900 validated synthetic topologies. Prior research on real-world core optical networks has been characterised by fragmented open data sources and disparate individual studies. Moreover, previous efforts have notably failed to provide synthetic data at a scale comparable to our present study. Topology Bench addresses this limitation, offering a unified resource and represents a 61.5% increase in spatially-referenced real world optical networks. To benchmark and identify the fundamental nature of optical network topologies through the lens of graph-theoretical analysis, we analyse both real and synthetic networks using structural, spatial and spectral metrics. Our comparative analysis identifies constraints in real optical network diversity and illustrates how synthetic networks can complement and expand the range of topologies available for use. Currently, topologies are selected based on subjective criteria, such as preference, data availability, or perceived suitability, leading to potential biases and limited representativeness. Our framework enhances the generalisability of optical network research by providing a more objective and systematic approach to topology selection. A statistical and correlation analysis reveals the quantitative range of all of these graph metrics and the relationships between them. Finally, we apply unsupervised machine learning to cluster real-world topologies into distinctive groups using nine optimal graph metrics using K-means. We conclude the analysis by providing guidance on how to use such clusters to select a diverse set of topologies for future studies.

cs.NI↗
arXiv2023

Multi-User Entanglement Distribution in Quantum Networks Using Multipath Routing

Quantum networks facilitate numerous applications such as secure communication and distributed quantum computation by performing entanglement distribution. Multi-user quantum applications where quantum information is shared between multiple users require access to a shared multipartite state between the users. We consider the problem of designing protocols for distributing such states, at an increased entanglement rate. We propose three protocols that increase the entanglement rate of multi-user applications by leveraging multipath routing. The protocols are evaluated on quantum networks with NISQ constraints, including limited quantum memories and probabilistic entanglement generation. Monte Carlo simulation results show that the developed protocols achieve an exponential speedup of entanglement rate compared to single-path routing techniques, with a maximum speedup of four orders of magnitude for the cases studied. The speedup was also found to improve for larger sets of users. When the protocols were tested in scaled-down real-world topologies, it was found that topology can have a significant effect on the achievable entanglement rates, with one order of magnitude difference between topologies. Finally, we find that the benefits of multipath routing are a maximum for short quantum memory decoherence times, and intermediate values of entanglement generation probability. Hence the protocols developed can benefit NISQ quantum network control and design.

quant-ph↗
arXiv2022

Resource Allocation in Multicore Elastic Optical Networks: A Deep Reinforcement Learning Approach

A deep reinforcement learning approach is applied, for the first time, to solve the routing, modulation, spectrum and core allocation (RMSCA) problem in dynamic multicore fiber elastic optical networks (MCF-EONs). To do so, a new environment - compatible with OpenAI's Gym - was designed and implemented to emulate the operation of MCF-EONs. The new environment processes the agent actions (selection of route, core and spectrum slot) by considering the network state and physical-layer-related aspects. The latter includes the available modulation formats and their reach and the inter-core crosstalk (XT), an MCF-related impairment. If the resulting quality of the signal is acceptable, the environment allocates the resources selected by the agent. After processing the agent's action, the environment is configured to give the agent a numerical reward and information about the new network state. The blocking performance of four different agents was compared through simulation to 3 baseline heuristics used in MCF-EONs. Results obtained for the NSFNet and COST239 network topologies show that the best-performing agent achieves, on average, up to a four-times decrease in blocking probability concerning the best-performing baseline heuristic methods.

cs.LG↗