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:2505.12764 · IEEE International Conference on Quantum Computing and Engineering (QCE)

Solving Graph Problems Using Permutation-Invariant Quantum Machine Learning

Maximilian Balthasar Mansky, Tobias Rohe, Gerhard Stenzel, Leo Sünkel, Alejandro Bravo de la Serna, Santiago Londoño Castillo, Gautham Sathish, Dimitra Nikolaidou, Dmytro Bondarenko, Linus Menzel, Claudia Linnhoff–Popien

Published Aug 30, 2025• 1 citation
arXivHTMLPDF

Abstract

Many computational problems are unchanged under some symmetry operation. In classical machine learning, this can be reflected with the layer structure of the neural network. In quantum machine learning, the ansatz can be tuned to correspond to the specific symmetry of the problem. We investigate this adaption of the quantum circuit to the problem symmetry on graph classification problems. On random graphs, the quantum machine learning ansatz classifies whether a given random graph is connected, bipartite, contains a Hamiltonian path or cycle, respectively. We find that if the quantum circuit reflects the inherent symmetry of the problem, it vastly outperforms the standard, unsymmetrized ansatzes. Even when the symmetry is only approximative, there is still a significant performance gain over non-symmetrized ansatzes. We show how the symmetry can be included in the quantum circuit in a straightforward constructive method.

DOI
10.1109/qce65121.2025.00178

Related papers

Equivariant quantum circuits for learning on weighted graphs

Andrea Skolik, Michele Cattelan, Sheir Yarkoni +2 more

Abstract Variational quantum algorithms are the leading candidate for advantage on near-term quantum hardware. When training a parametrized quantum circuit in this setting to solve a specific problem, the choice of ansatz is one of the most important factors that determines the trainability and performance of the algorithm. In quantum machine learning (QML), however, the literature on ansatzes that are motivated by the training data structure is scarce. In this work, we introduce an ansatz for learning tasks on weighted graphs that respects an important graph symmetry, namely equivariance under node permutations. We evaluate the performance of this ansatz on a complex learning task, namely neural combinatorial optimization, where a machine learning model is used to learn a heuristic for a combinatorial optimization problem. We analytically and numerically study the performance of our model, and our results strengthen the notion that symmetry-preserving ansatzes are a key to success in QML.

arXiv:2205.06109•npj Quantum Information•May 13, 2023•71 citations

Clique Detection Using Symmetry-Restricted Quantum Circuits

Maximilian Balthasar Mansky, Tobias Rohe, Jonas Stein +3 more

We show the application of permutation-invariant quantum circuits to the clique problem. The experiment asks to label a clique through identification of the nodes in a larger subgraph. The permutation-invariant quantum circuit outperforms a cyclic-invariant alternative as well as a standard quantum machine learning ansatz. We explain the behavior through the intrinsic symmetry of the problem, in the sense that the problem is symmetric under permutation of both the feature and the label.

arXiv:2506.03339•IEEE International Conference on Quantum Computing and Engineering (QCE)•Aug 30, 2025•0 citations

Analysis of quantum neural network performance via edge cases

Maximilian Balthasar Mansky, Tobias Rohe, Linus Menzel +2 more

We evaluate the particular performance of different quantum machine learning networks on a graph classification task. Quantum circuits with varying internal symmetry that completely, partially and not at all confer to the symmetry of the graph show different performance on the data set. The convergence results are inspected using a number of special graphs with particular structure. These are unlikely to occur in the training data and cover specific cases that refute the assumption that the quantum neural network learns simpler surrogate models based on the number of edges in the graph.

arXiv:2506.12427•Jun 14, 2025•0 citations

Compositional Quantum Heuristics for Max-Clique Detection

Tiffany Duneau, Colin Krawchuk, Anna Pearson

Quantum machine learning holds the promise of combining the success of classical machine learning methods with the power of quantum computing, however one of the largest obstacles facing the field is the problem of barren plateaus. Parameterised quantum circuits offer a flexible framework for developing quantum machine learning models, but their practicality is constrained by a trade-off between trainability and classical simulability. In general, circuits that are sufficiently expressive to model complex behaviour often exhibit barren plateaus, where gradients vanish and optimisation fails. In this work we investigate a compositional approach to mitigate this trade-off by assembling larger quantum models from smaller subcomponents. To ensure trainability of these subcomponents, we describe a framework for constructing group-invariant loss functions, which introduce symmetry-induced inductive bias and lead to improved gradient behaviour and generalisation. In particular, we use this framework to design permutation-equivariant quantum graph neural networks for identifying maximal cliques in graphs. The models we construct exhibit superior training gradients through symmetry-induced bias, and our experiments demonstrate that the trained models generalise to larger, more complex problem instances. Finally, inspired by Quantum-Informed Recursive Optimisation Algorithms (arXiv:2308.13607), we implement a recursive hybrid quantum-classical heuristic using the learned quantum models to guide a classical search procedure, demonstrating improved inference accuracy and scalability. Together, these results suggest that compositional circuits could be a viable pathway towards scalable quantum learning models that remain challenging to reproduce classically.

quant-pharXiv:2605.07611•Quantum Science and Technology•May 11, 2026•0 citations

Machine learning applied to the quantum circuit layout problem

André Filipe Mestre Mendes

This dissertation presents an innovative approach to the Quantum Circuit Layout problem by leveraging advanced Machine Learning techniques, specifically Graph Neural Networks. The Quantum Circuit Layout problem is crucial for optimizing the placement and routing of qubits in quantum circuits, particularly under the constraints of current quantum hardware. By establishing graph representations for the quantum circuits and computers, we provide a foundation for understanding the complexity of the Quantum Circuit Layout problem, demonstrating its equivalence to the well-studied Subgraph Isomorphism problem, which is known to be NP-Complete. Our proposed solution, the Adaptive Edge-Deleting Network, employs a novel adaptive edgedeleting mechanism to enhance the extraction of relevant features from these graphs while maintaining the structural integrity essential for subgraph matching. By iteratively refining node representations through neural message passing and incorporating a unique attention mechanism, Adaptive Edge-Deleting Network significantly improves the efficiency of graph embedding allowing for a novel way to solve the Quantum Circuit Layout problem.

•Apr 1, 2025•0 citations

Ideal Parametrisation Estimation for Variational Quantum Circuit Classifiers Using Machine Learning

Mostafa Fathi, Walid E. Gomaa, Yasutaka Wada +3 more

Quantum neural network implementation through variational quantum circuits is getting attention due to their applicability to the existing quantum computer technology and significant model parameters reduction. This paper considers the problem of automatically transforming existing neural networks into corresponding variational quantum circuits. In particular, given a trained neural network, the paper considers estimating the number of ansatz layers so that the corresponding circuit is not over or under-parametrized. This is achieved by constructing a dataset containing many trained neural networks each labelled with the corresponding optimal number of variational quantum layers, obtained through experimentation. The paper further performs dataset augmentation using an LLM model to speed up dataset construction. The paper then selects a set of features based on both the network architecture and the trained parameters, and then trains a feed-forward neural network. The trained network achieves 86.7% accuracy on unseen unaugmented test circuits.

IEEE International Conference on Quantum Computing and Engineering (QCE)•Sep 15, 2024•0 citations
Many computational problems are unchanged under some symmetry operation. In classical machine learning, this can be reflected with the layer structure of the neural network. In quantum machine learning, the ansatz can be tuned to correspond to the specific symmetry of the problem. We investigate this adaption of the quantum circuit to the problem symmetry on graph classification problems. On random graphs, the quantum machine learning ansatz classifies whether a given random graph is connected, bipartite, contains a Hamiltonian path or cycle, respectively. We find that if the quantum circuit reflects the inherent symmetry of the problem, it vastly outperforms the standard, unsymmetrized ansatzes. Even when the symmetry is only approximative, there is still a significant performance gain over non-symmetrized ansatzes. We show how the symmetry can be included in the quantum circuit in a straightforward constructive method.
Abstract Variational quantum algorithms are the leading candidate for advantage on near-term quantum hardware. When training a parametrized quantum circuit in this setting to solve a specific problem, the choice of ansatz is one of the most important factors that determines the trainability and performance of the algorithm. In quantum machine learning (QML), however, the literature on ansatzes that are motivated by the training data structure is scarce. In this work, we introduce an ansatz for learning tasks on weighted graphs that respects an important graph symmetry, namely equivariance under node permutations. We evaluate the performance of this ansatz on a complex learning task, namely neural combinatorial optimization, where a machine learning model is used to learn a heuristic for a combinatorial optimization problem. We analytically and numerically study the performance of our model, and our results strengthen the notion that symmetry-preserving ansatzes are a key to success in QML.
We evaluate the particular performance of different quantum machine learning networks on a graph classification task. Quantum circuits with varying internal symmetry that completely, partially and not at all confer to the symmetry of the graph show different performance on the data set. The convergence results are inspected using a number of special graphs with particular structure. These are unlikely to occur in the training data and cover specific cases that refute the assumption that the quantum neural network learns simpler surrogate models based on the number of edges in the graph.
Quantum machine learning holds the promise of combining the success of classical machine learning methods with the power of quantum computing, however one of the largest obstacles facing the field is the problem of barren plateaus. Parameterised quantum circuits offer a flexible framework for developing quantum machine learning models, but their practicality is constrained by a trade-off between trainability and classical simulability. In general, circuits that are sufficiently expressive to model complex behaviour often exhibit barren plateaus, where gradients vanish and optimisation fails. In this work we investigate a compositional approach to mitigate this trade-off by assembling larger quantum models from smaller subcomponents. To ensure trainability of these subcomponents, we describe a framework for constructing group-invariant loss functions, which introduce symmetry-induced inductive bias and lead to improved gradient behaviour and generalisation. In particular, we use this framework to design permutation-equivariant quantum graph neural networks for identifying maximal cliques in graphs. The models we construct exhibit superior training gradients through symmetry-induced bias, and our experiments demonstrate that the trained models generalise to larger, more complex problem instances. Finally, inspired by Quantum-Informed Recursive Optimisation Algorithms (arXiv:2308.13607), we implement a recursive hybrid quantum-classical heuristic using the learned quantum models to guide a classical search procedure, demonstrating improved inference accuracy and scalability. Together, these results suggest that compositional circuits could be a viable pathway towards scalable quantum learning models that remain challenging to reproduce classically.
This dissertation presents an innovative approach to the Quantum Circuit Layout problem by leveraging advanced Machine Learning techniques, specifically Graph Neural Networks. The Quantum Circuit Layout problem is crucial for optimizing the placement and routing of qubits in quantum circuits, particularly under the constraints of current quantum hardware. By establishing graph representations for the quantum circuits and computers, we provide a foundation for understanding the complexity of the Quantum Circuit Layout problem, demonstrating its equivalence to the well-studied Subgraph Isomorphism problem, which is known to be NP-Complete. Our proposed solution, the Adaptive Edge-Deleting Network, employs a novel adaptive edgedeleting mechanism to enhance the extraction of relevant features from these graphs while maintaining the structural integrity essential for subgraph matching. By iteratively refining node representations through neural message passing and incorporating a unique attention mechanism, Adaptive Edge-Deleting Network significantly improves the efficiency of graph embedding allowing for a novel way to solve the Quantum Circuit Layout problem.
Quantum neural network implementation through variational quantum circuits is getting attention due to their applicability to the existing quantum computer technology and significant model parameters reduction. This paper considers the problem of automatically transforming existing neural networks into corresponding variational quantum circuits. In particular, given a trained neural network, the paper considers estimating the number of ansatz layers so that the corresponding circuit is not over or under-parametrized. This is achieved by constructing a dataset containing many trained neural networks each labelled with the corresponding optimal number of variational quantum layers, obtained through experimentation. The paper further performs dataset augmentation using an LLM model to speed up dataset construction. The paper then selects a set of features based on both the network architecture and the trained parameters, and then trains a feed-forward neural network. The trained network achieves 86.7% accuracy on unseen unaugmented test circuits.