Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2510.01918 · IEEE International Conference on Quantum Artificial Intelligence (QAI)

Hybrid Quantum-Classical Walks for Graph Representation Learning in Community Detection

A. J. Soriano Marin, Mauricio Soto-Gomez, Giorgio Valentini, Elena Casiraghi, Carlos Cano, Daniel Manzano

Published Nov 2, 2025• 2 citations
arXivHTMLPDF

Abstract

Graph Representation Learning (GRL) has emerged as a cornerstone technique for analysing complex, networked data across diverse domains, including biological systems, social networks, and data analysis. Traditional GRL methods often struggle to capture intricate relationships within complex graphs, particularly those exhibiting non-trivial structural properties such as power-law distributions or hierarchical structures. This paper introduces a novel quantum-inspired algorithm for GRL, utilizing hybrid Quantum-Classical Walks to overcome these limitations. Our approach combines the benefits of both quantum and classical dynamics, allowing the walker to simultaneously explore both highly local and far-reaching connections within the graph. Preliminary results for a case study in network community detection shows that this hybrid dynamic enables the algorithm to adapt effectively to complex graph topologies, offering a robust and versatile solution for GRL tasks.

DOI
10.1109/qai63978.2025.00016

Related papers

Discrete quantum-classical walks for link prediction

A. Marin, M. Soto-Gomez, Giorgio Valentini +3 more

Link prediction is a fundamental task in network science that involves estimating the likelihood of missing or future connections between nodes based on the structural patterns of a graph. While Graph Neural Networks (GNNs) have become a dominant approach, Random Walk (RW)-based methods remain highly relevant due to their lower computational demands in large-scale networks. Although recent hybrid quantum-classical walks have shown promise in exploring graph structures, existing continuous-time models are computationally expensive as they require solving high-dimensional systems of differential equations. In this paper, we propose a discrete version of a recent hybrid quantum-classical random walk algorithm that significantly improves computational efficiency by utilizing state vector representations instead of density matrices. Our method, inspired by the Quantum Jumps approach, alternates between coherent evolution and measurement-driven state collapse to generate trajectories for node embeddings. We integrate this algorithm into a link prediction pipeline and evaluate its performance across various synthetic topologies, including Grid, Power-Law, Erdös-Rényi, Barabási-Albert, and Community graphs. The experimental results demonstrate that the proposed discrete hybrid walks consistently outperform classical random walk approaches in Community and Grid topologies, showing that quantum-inspired dynamics can capture complex structural information more effectively than classical dynamics for specific topologies.

IEEE Conference on Artificial Intelligence (CAI)•May 8, 2026•0 citations

Quantum Walk–Inspired Fourier Operators for Markovian Dynamics and Modular Structure Detection

Yanfei Ma, Daozheng Qu, Yibo Wang

Community recognition continues to be a significant difficulty in the analysis of complex networks, especially in dynamic and stochastic contexts where communities change over time. Conventional spectral clustering and Markovian models either include structural regularities or temporal dependencies, although frequently lack the ability to amalgamate both within a cohesive theoretical framework. This study presents a Fourier operator framework inspired by quantum walks that integrates spectral graph theory, Markov processes, and quantum brain dynamics for the discovery of modular structures. We specifically build Fourier-Markov operators based on quantum walk principles to encapsulate both global frequency components and local transition dynamics of evolving networks. Our system utilizes superposition and interference effects by mapping random walks into a quantum-inspired Hilbert space, therefore uncovering latent modular structures with enhanced resolution. The resultant operator family facilitates interpretable community partitions, scalable computation using hybrid classicalquantum methodologies, and resilient detection across temporal snapshots. Experimental assessments on benchmark dynamic networks—encompassing social, biological, and financial graphs—indicate that our approach attains enhanced modularity optimization, temporal stability, and uncertainty calibration relative to leading community detection benchmarks. This study emphasizes the promise of quantum-spectral operators as a novel basis for examining Markovian community dynamics in intricate systems.

International Conference on Machine Learning and Computer Application (ICMLCA)•Oct 17, 2025•0 citations

Quantum Graph Embeddings for Complex Data Analysis

Jincheng Zhang

Quantum Graph Embeddings represent a novel approach to data analysis, leveraging the power of quantum algorithms to create and manipulate graph structures. This paper introduces a method for generating quantum graphs with enhanced representation capabilities, specifically designed for complex data analysis. We explore the potential of quantum algorithms to efficiently explore graph structure, identify key nodes and connections, and ultimately uncover hidden patterns and relationships within datasets. The core mechanism focuses on creating a scalable and adaptable framework for data representation, addressing limitations inherent in traditional graph embedding techniques. This work proposes a method that is inherently scalable, potentially offering significant advantages in the analysis of intricate and high-dimensional datasets.

•Sep 4, 2026•0 citations

基于量子自适应的图神经网络

Jincheng Zhang

This paper introduces a novel approach to graph neural networks (GNNs) – the 'Quantum Adaptive Graph Neural Network' – leveraging quantum self-adaptation to enhance both expressiveness and computational efficiency. Traditional GNNs often struggle with the inherent complexities of graph data and require substantial computational resources. We propose a new architecture that employs quantum mechanics to simulate graph structures, allowing the network to learn more robust representations. This approach, dubbed Quantum Adaptive Graph Neural Network (QAGNN), utilizes quantum machine learning algorithms to improve the network's ability to capture intricate relationships within the graph. The paper details the core mechanism, its mathematical foundations, and preliminary experimental results demonstrating significant improvements in various graph classification tasks compared to traditional GNNs. We present a comprehensive analysis of the QAGNN's performance and discuss potential future research directions.

•Aug 31, 2026•0 citations

Integrating Quantum Approximate Optimization Algorithm with Machine Learning for Graph-Structured Problem Solving

Dr. Mamta Singh

The integration of quantum algorithms into machine learning pipelines offers a promising direction for addressing complex combinatorial optimization problems inherent in graph-structured data. This study investigates the use of the Quantum Approximate Optimization Algorithm (QAOA) as a quantum subroutine within a hybrid quantum–classical framework to enhance machine learning solutions for graph-based tasks. QAOA is employed to solve NP-hard graph optimization problems such as Max-Cut, graph coloring, and community detection, which are reformulated as Quadratic Unconstrained Binary Optimization (QUBO) problems. The solutions provided by QAOA are then incorporated into machine learning models, such as Graph Neural Networks (GNNs) or clustering algorithms, as enriched features or optimized graph partitions. The hybrid approach is evaluated on benchmark datasets, demonstrating competitive accuracy and improved scalability in certain problem instances compared to purely classical methods. The findings highlight the potential of leveraging near-term quantum hardware for augmenting classical learning systems in domains such as logistics, social network analysis, and computational biology.

•Mar 25, 2026•0 citations

Community detection in network using Szegedy quantum walk

Md Samsur Rahaman, Supriyo Dutta

In a network, the vertices with similar characteristics construct communities. The vertices in a community are well-connected. Detecting the communities in a network is a challenging and important problem in the theory of complex networks. One approach to solving this problem uses the classical random walks on graphs. In quantum computing, quantum walks are the quantum mechanical counterparts of classical random walks. In this article, we employ a variant of Szegedy's quantum walk to develop a procedure for discovering the communities in networks. The limiting probability distribution of quantum walks assists us in determining the inclusion of a vertex in a community. We apply our community detection procedure to a variety of graphs and social networks, including the relaxed caveman graph, $l$-partition graph, Karate club graph, and the dolphin's social network, among others.

quant-pharXiv:2601.21152•Jan 30, 2026•1 citation
Graph Representation Learning (GRL) has emerged as a cornerstone technique for analysing complex, networked data across diverse domains, including biological systems, social networks, and data analysis. Traditional GRL methods often struggle to capture intricate relationships within complex graphs, particularly those exhibiting non-trivial structural properties such as power-law distributions or hierarchical structures. This paper introduces a novel quantum-inspired algorithm for GRL, utilizing hybrid Quantum-Classical Walks to overcome these limitations. Our approach combines the benefits of both quantum and classical dynamics, allowing the walker to simultaneously explore both highly local and far-reaching connections within the graph. Preliminary results for a case study in network community detection shows that this hybrid dynamic enables the algorithm to adapt effectively to complex graph topologies, offering a robust and versatile solution for GRL tasks.
Link prediction is a fundamental task in network science that involves estimating the likelihood of missing or future connections between nodes based on the structural patterns of a graph. While Graph Neural Networks (GNNs) have become a dominant approach, Random Walk (RW)-based methods remain highly relevant due to their lower computational demands in large-scale networks. Although recent hybrid quantum-classical walks have shown promise in exploring graph structures, existing continuous-time models are computationally expensive as they require solving high-dimensional systems of differential equations. In this paper, we propose a discrete version of a recent hybrid quantum-classical random walk algorithm that significantly improves computational efficiency by utilizing state vector representations instead of density matrices. Our method, inspired by the Quantum Jumps approach, alternates between coherent evolution and measurement-driven state collapse to generate trajectories for node embeddings. We integrate this algorithm into a link prediction pipeline and evaluate its performance across various synthetic topologies, including Grid, Power-Law, Erdös-Rényi, Barabási-Albert, and Community graphs. The experimental results demonstrate that the proposed discrete hybrid walks consistently outperform classical random walk approaches in Community and Grid topologies, showing that quantum-inspired dynamics can capture complex structural information more effectively than classical dynamics for specific topologies.
Community recognition continues to be a significant difficulty in the analysis of complex networks, especially in dynamic and stochastic contexts where communities change over time. Conventional spectral clustering and Markovian models either include structural regularities or temporal dependencies, although frequently lack the ability to amalgamate both within a cohesive theoretical framework. This study presents a Fourier operator framework inspired by quantum walks that integrates spectral graph theory, Markov processes, and quantum brain dynamics for the discovery of modular structures. We specifically build Fourier-Markov operators based on quantum walk principles to encapsulate both global frequency components and local transition dynamics of evolving networks. Our system utilizes superposition and interference effects by mapping random walks into a quantum-inspired Hilbert space, therefore uncovering latent modular structures with enhanced resolution. The resultant operator family facilitates interpretable community partitions, scalable computation using hybrid classicalquantum methodologies, and resilient detection across temporal snapshots. Experimental assessments on benchmark dynamic networks—encompassing social, biological, and financial graphs—indicate that our approach attains enhanced modularity optimization, temporal stability, and uncertainty calibration relative to leading community detection benchmarks. This study emphasizes the promise of quantum-spectral operators as a novel basis for examining Markovian community dynamics in intricate systems.
Quantum Graph Embeddings represent a novel approach to data analysis, leveraging the power of quantum algorithms to create and manipulate graph structures. This paper introduces a method for generating quantum graphs with enhanced representation capabilities, specifically designed for complex data analysis. We explore the potential of quantum algorithms to efficiently explore graph structure, identify key nodes and connections, and ultimately uncover hidden patterns and relationships within datasets. The core mechanism focuses on creating a scalable and adaptable framework for data representation, addressing limitations inherent in traditional graph embedding techniques. This work proposes a method that is inherently scalable, potentially offering significant advantages in the analysis of intricate and high-dimensional datasets.
This paper introduces a novel approach to graph neural networks (GNNs) – the 'Quantum Adaptive Graph Neural Network' – leveraging quantum self-adaptation to enhance both expressiveness and computational efficiency. Traditional GNNs often struggle with the inherent complexities of graph data and require substantial computational resources. We propose a new architecture that employs quantum mechanics to simulate graph structures, allowing the network to learn more robust representations. This approach, dubbed Quantum Adaptive Graph Neural Network (QAGNN), utilizes quantum machine learning algorithms to improve the network's ability to capture intricate relationships within the graph. The paper details the core mechanism, its mathematical foundations, and preliminary experimental results demonstrating significant improvements in various graph classification tasks compared to traditional GNNs. We present a comprehensive analysis of the QAGNN's performance and discuss potential future research directions.
The integration of quantum algorithms into machine learning pipelines offers a promising direction for addressing complex combinatorial optimization problems inherent in graph-structured data. This study investigates the use of the Quantum Approximate Optimization Algorithm (QAOA) as a quantum subroutine within a hybrid quantum–classical framework to enhance machine learning solutions for graph-based tasks. QAOA is employed to solve NP-hard graph optimization problems such as Max-Cut, graph coloring, and community detection, which are reformulated as Quadratic Unconstrained Binary Optimization (QUBO) problems. The solutions provided by QAOA are then incorporated into machine learning models, such as Graph Neural Networks (GNNs) or clustering algorithms, as enriched features or optimized graph partitions. The hybrid approach is evaluated on benchmark datasets, demonstrating competitive accuracy and improved scalability in certain problem instances compared to purely classical methods. The findings highlight the potential of leveraging near-term quantum hardware for augmenting classical learning systems in domains such as logistics, social network analysis, and computational biology.
In a network, the vertices with similar characteristics construct communities. The vertices in a community are well-connected. Detecting the communities in a network is a challenging and important problem in the theory of complex networks. One approach to solving this problem uses the classical random walks on graphs. In quantum computing, quantum walks are the quantum mechanical counterparts of classical random walks. In this article, we employ a variant of Szegedy's quantum walk to develop a procedure for discovering the communities in networks. The limiting probability distribution of quantum walks assists us in determining the inclusion of a vertex in a community. We apply our community detection procedure to a variety of graphs and social networks, including the relaxed caveman graph, lll-partition graph, Karate club graph, and the dolphin's social network, among others.