arXiv subjects

Mehdi Djellabi

Publications and source records attributed to Mehdi Djellabi.

7 recordsLinked to original sources

arXiv2026

On the Expressive Power of the Transverse-Field Ising Model for Graph Learning

We study the quantum evolution induced by graph-indexed Ising Hamiltonians as a source of structural signal for graph learning. Graph automorphisms preserve symmetries of the Hamiltonian, and these symmetries constrain the quantum evolution in a way that turns time-dependent local measurements into informative probes of graph structure. Leveraging this idea, we introduce QDAGer, a quantum-inspired graph-pair Transformer that injects quantum-dynamical features from time series of node occupations and connected two-point correlators directly into the attention mechanism. We apply QDAGer to learning Graph Edit Distance (GED), an NP-hard similarity measure, using either a direct permutation-invariant embedding discrepancy or an alignment-based surrogate loss. Experiments on multiple GED benchmarks under different edit cost settings show that the proposed dynamical features provide a stronger inductive bias than classical structural alternatives under the same training protocol. In addition, we report ablations where the dynamical signal is replaced by standard random-walk and heat-kernel features while keeping the architecture fixed, highlighting that the gain comes from the injected dynamics rather than model capacity alone.

quant-ph↗

arXiv2026

Attributed-graphs kernel implementation using local detuning of neutral-atoms Rydberg Hamiltonian

We extend the quantum-feature kernel framework, which relies on measurements of graph-dependent observables, along three directions. First, leveraging neutral-atom quantum processing units (QPUs), we introduce a scheme that incorporates attributed graphs by embedding edge features into atomic positions and node features into local detuning fields of a Rydberg Hamiltonian. We demonstrate both theoretically and empirically that local detuning enhances kernel expressiveness. Second, in addition to the existing quantum evolution kernel (QEK), which uses global observables, we propose the generalized-distance quantum-correlation (GDQC) kernel, based on local observables. While the two kernels show comparable performance, we show that GDQC can achieve higher expressiveness. Third, instead of restricting to observables at single time steps, we combine information from multiple stages of the quantum evolution via pooling operations. Using extensive simulations on two molecular benchmark datasets, MUTAG and PTC\_FM, we find: (a) QEK and GDQC perform competitively with leading classical algorithms; and (b) pooling further improves performance, enabling quantum-feature kernels to surpass classical baselines. These results show that node-feature embedding and kernel designs based on local observables advance quantum-enhanced graph machine learning on neutral-atom devices.

quant-ph↗

arXiv2025

Quantum Graph Attention Networks: Trainable Quantum Encoders for Inductive Graph Learning

We introduce Quantum Graph Attention Networks (QGATs) as trainable quantum encoders for inductive learning on graphs, extending the Quantum Graph Neural Networks (QGNN) framework. QGATs leverage parameterized quantum circuits to encode node features and neighborhood structures, with quantum attention mechanisms modulating the contribution of each neighbor via dynamically learned unitaries. This allows for expressive, locality-aware quantum representations that can generalize across unseen graph instances. We evaluate our approach on the QM9 dataset, targeting the prediction of various chemical properties. Our experiments compare classical and quantum graph neural networks-with and without attention layers-demonstrating that attention consistently improves performance in both paradigms. Notably, we observe that quantum attention yields increasing benefits as graph size grows, with QGATs significantly outperforming their non-attentive quantum counterparts on larger molecular graphs. Furthermore, for smaller graphs, QGATs achieve predictive accuracy comparable to classical GAT models, highlighting their viability as expressive quantum encoders. These results show the potential of quantum attention mechanisms to enhance the inductive capacity of QGNN in chemistry and beyond.

quant-ph↗

arXiv2024

Quantum Positional Encodings for Graph Neural Networks

In this work, we propose novel families of positional encodings tailored to graph neural networks obtained with quantum computers. These encodings leverage the long-range correlations inherent in quantum systems that arise from mapping the topology of a graph onto interactions between qubits in a quantum computer. Our inspiration stems from the recent advancements in quantum processing units, which offer computational capabilities beyond the reach of classical hardware. We prove that some of these quantum features are theoretically more expressive for certain graphs than the commonly used relative random walk probabilities. Empirically, we show that the performance of state-of-the-art models can be improved on standard benchmarks and large-scale datasets by computing tractable versions of quantum features. Our findings highlight the potential of leveraging quantum computing capabilities to enhance the performance of transformers in handling graph data.

quant-ph↗

arXiv2024

Graph Algorithms with Neutral Atom Quantum Processors

Neutral atom technology has steadily demonstrated significant theoretical and experimental advancements, positioning itself as a front-runner platform for running quantum algorithms. One unique advantage of this technology lies in the ability to reconfigure the geometry of the qubit register, from shot to shot. This unique feature makes possible the native embedding of graph-structured problems at the hardware level, with profound consequences for the resolution of complex optimization and machine learning tasks. By driving qubits, one can generate processed quantum states which retain graph complex properties. These states can then be leveraged to offer direct solutions to problems or as resources in hybrid quantum-classical schemes. In this paper, we review the advancements in quantum algorithms for graph problems running on neutral atom Quantum Processing Units (QPUs), and discuss recently introduced embedding and problem-solving techniques. In addition, we clarify ongoing advancements in hardware, with an emphasis on enhancing the scalability, controllability and computation repetition rate of neutral atom QPUs.

quant-ph↗

arXiv2023

Enhancing Graph Neural Networks with Quantum Computed Encodings

Transformers are increasingly employed for graph data, demonstrating competitive performance in diverse tasks. To incorporate graph information into these models, it is essential to enhance node and edge features with positional encodings. In this work, we propose novel families of positional encodings tailored for graph transformers. These encodings leverage the long-range correlations inherent in quantum systems, which arise from mapping the topology of a graph onto interactions between qubits in a quantum computer. Our inspiration stems from the recent advancements in quantum processing units, which offer computational capabilities beyond the reach of classical hardware. We prove that some of these quantum features are theoretically more expressive for certain graphs than the commonly used relative random walk probabilities. Empirically, we show that the performance of state-of-the-art models can be improved on standard benchmarks and large-scale datasets by computing tractable versions of quantum features. Our findings highlight the potential of leveraging quantum computing capabilities to potentially enhance the performance of transformers in handling graph data.

quant-ph↗

arXiv2021

Clustering of temporal nodes profiles in dynamic networks of contacts

Stream graphs are a very useful mode of representation for temporal network data, whose richness offers a wide range of possible approaches. The various methods aimed at generalising the classical approaches applied to static networks are constantly being improved. In this paper, we describe a framework that extend to stream graphs iterative weighted-rich-clubs characterisation for static networks proposed in [1]. The general principle is that we no longer consider the membership of a node to one of the weighted-rich-clubs for the whole time period, but each node is associated with a temporal profile which is the concatenation of the successive memberships of the node to the weighted-rich-clubs that appear, disappear and change all along the period. A clustering of these profiles gives the possibility to establish a reduced list of typical temporal profiles and so a more in-depth understanding of the temporal structure of the network. This approach is tested on real world data produced by recording the interactions between different students within their respective schools. [1] M. Djellabi, B. Jouve, and F. Amblard. Dense and sparse vertex connectivity in networks. Journal of Complex Networks, 8(3), 2020.

cs.SI↗