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:2609.15139

A train--prune--readout--rewrite workflow for interpretable quantum learning

Siran Zhang, Shuming Cheng, Xiang Li, Jinyi Liu

quant-phSubmitted Sep 14, 2026• Announced Sep 15, 2026• 0 citations
arXivHTMLPDF

Abstract

AI for Science aims not only to predict complex physical systems from data, but also to extract mathematical structure and physically testable representations from learned models. Here, a train--prune--readout--rewrite workflow is developed that separates physical-domain grounding from three increasingly stringent analysis claims: algebraically equivalent readout of a trained predictor, compact teacher-faithful symbolic rewriting on the sampled physical domain, and transformation-based tests of learned internal representations. The workflow is implemented with complex-valued Kolmogorov--Arnold networks, whose explicit edge functions enable post-pruning analytic readout of the retained computation. In analytically controlled single-qubit tasks, rewriting recovered the quadratic structure of purity, whereas von Neumann entropy yielded only a domain-bounded symbolic surrogate; physics-aligned variable grouping preserved symbolic fidelity. For two-qubit entanglement-related tasks, shared learning exposed a common internal representation whose physical content was interrogated directly. Local-unitary transformations rejected a direct invariant-coordinate interpretation, while fixed-decoder transfer showed that the shared activation carries Pauli-correlation information in a transformation-consistent form. Task-related invariant spectral features were subsequently recovered through low-order nonlinear readouts. Separate predictive tests retained high accuracy for three-qubit classification and controlled ten-qubit purity regression with over one million complex inputs. These results establish an evidence-resolved framework for distinguishing physical grounding, readable computation, faithful symbolic compression and transformation-tested physical structure in constrained complex-valued scientific learning.

Related papers

Sparsified Kolmogorov-Arnold Networks for Interpretable Quantum State Tomography

Xinge Wu, Huaxin Wang, Jiajun Liu +4 more

Machine-learning approaches to quantum state tomography can achieve high reconstruction fidelity, but the physical structure used by the trained model often remains implicit. Here we ask whether a sparsified Kolmogorov-Arnold Network (KAN) can be used not only as a regressor, but also as an inspectable reconstruction rule whose internal organization can be checked against known Pauli structure. We study a controlled three-qubit GHZ-family benchmark in which all 63 non-identity Pauli expectation values are used to reconstruct three GHZ-subspace variables: the population imbalance $z$, the real off-diagonal component $c$, and the imaginary off-diagonal component $s$. Under finite-shot sampling and depolarizing noise, external ablation identifies the extended 12-channel GHZ-relevant Pauli set from the 63 measurements, with exact top-12 recovery across the tested shot counts and depolarizing-noise strengths. These support patterns remain stable across multi-seed random-initialization and noise-level analyses, and collapse under random-label controls. The dominant pruned input-hidden-output pathways organize Z-type population observables and X/Y off-diagonal observables in a pattern consistent with the analytic GHZ Pauli grouping, and sparse formula recovery recovers the canonical signed Pauli relations. The contribution of the KAN is therefore pathway-level structural interpretability within a neural reconstruction model, rather than superior sparse regression. Together with negative controls, these probes provide a consistency chain for auditing learned reconstruction rules against known physical structure.

quant-pharXiv:2606.11814•Jun 11, 2026•0 citations

Adaptive Quantum Learning Architectures with Superalgebraic Entangled Layers

Murali Krishna Pasupuleti

This paper introduces a novel quantum machine learning framework that integrates adaptive learning capabilities with superalgebraically structured entangled layers. Leveraging the mathematical formalism of Lie superalgebras—specifically osp(1∣2) and sl(2∣1)—the proposed architecture encodes quantum states in graded Hilbert spaces and governs their evolution via symmetry-preserving supercharges. The entangled layers dynamically adapt to task-specific feedback through a hybrid quantum-classical training protocol that ensures algebraic consistency and quantum coherence. Symbolic computation and quantum simulation tools are employed to validate the fidelity and symmetry of entanglement evolution. The system is benchmarked on tasks such as symbolic reasoning, quantum phase classification, and structured regression, outperforming baseline models in interpretability, stability, and generalization. This research lays the foundation for scalable, physically grounded quantum AI systems capable of learning through algebraic alignment with fundamental symmetries. Keywords: quantum machine learning, adaptive learning, entanglement, superalgebra, osp(1∣2), graded Hilbert space, quantum AI, symmetry-preserving learning, Lie superalgebras, supercharges, hybrid quantum-classical architecture, interpretable QML, symbolic reasoning, quantum circuits

International Journal of Academic and Industrial Research Innovations(IJAIRI)•Apr 11, 2025•0 citations

Quantum Orthogonal Separable Physics-Informed Neural Networks

Pietro Zanotta, Ljubomir Budinski, Caglar Aytekin +1 more

We introduce Quantum Orthogonal Separable Physics-Informed Neural Networks (QO-SPINNs), which replace selected dense hidden transformations in physics-informed PDE models with quantum-compatible orthogonal maps. A pyramidal reconfigurable-beamsplitter circuit realizes these maps with $Θ(d)$ circuit depth. Under the stated signed-readout access model and fixed per-call failure probability, a width-$d$ map has sequential access-cost upper bound $\mathcal O(d\log d/ε^2)$, which is asymptotically smaller than the $Θ(d^2)$ cost of conventional direct dense multiplication when $ε(d)^{-2}\log d=o(d)$. Because unary state preparation discards magnitude, we retain each hidden-state norm classically and restore it after readout, yielding the exact map $Wh$ under ideal readout. We study this construction in two complementary settings: forward PDE regression and a norm-controlled residual model for representation geometry and uncertainty quantification. In the norm-controlled model, structural orthogonality yields explicit bi-Lipschitz bounds, while finite-readout analysis relates sufficient reconstruction precision to a prescribed physical-coordinate resolution through trained-model constants. This makes the access-cost condition assessable for trained checkpoints, although favorable fixed-resolution scaling across widths is not established. A post-hoc Bayesian-linear covariance provides uncertainty from the same frozen predictor, and cross-PDE transfer shows that strong distance-awareness can persist even when absolute coverage calibration fails. All Quantum Orthogonal (QO) layers are evaluated by classical simulation: the results therefore characterize a quantum-compatible structured neural primitive rather than demonstrating a practical quantum-hardware advantage.

quant-pharXiv:2511.12613•Nov 18, 2025•0 citations

Quantum–AI for High-Dimensional Learning: Graph Neural Networks, Topological Data Analysis, Tensor Networks, and Sparse Predictive Modelling

Murali Krishna Pasupuleti

Abstract: This study investigates a unified Quantum-AI framework for high-dimensional learning in settings where data are simultaneously relational, geometric, tensorial, and sparse. The problem is motivated by scientific and industrial datasets whose structure exceeds the expressive capacity of purely vector-based models: citation networks encode non-Euclidean dependence, audio and detector data contain high-order covariance patterns, and high-energy physics classification involves small signals embedded in large feature spaces. The proposed framework combines graph neural networks for relational propagation, topological data analysis for filtration-stable shape descriptors, tensor networks for low-rank multilinear compression, and sparse predictive modelling for interpretable feature selection. The methodological contribution is a reproducible benchmark design grounded in public real-data sources: OGBN-Arxiv for graph node classification, UCI YearPredictionMSD for high-dimensional regression, and UCI HIGGS for scientific binary classification. Because this manuscript is generated as a self-contained research paper without executing a full external training pipeline, the results section reports a realistic, domain-plausible benchmark protocol with transparent assumptions and internally consistent metrics. Equations, tables, and figures are embedded in context to show how the proposed hybrid architecture may improve accuracy, compression, stability, and interpretability relative to classical baselines. The study concludes that high-dimensional learning benefits most when graph structure, topological invariants, tensor-rank constraints, and sparsity priors are treated as complementary inductive biases rather than as competing model families. Keywords Quantum artificial intelligence; high-dimensional learning; graph neural networks; topological data analysis; tensor networks; sparse predictive modelling; persistent homology; OGBN-Arxiv; HIGGS; YearPredictionMSD; quantum kernels; variational circuits; scalable machine learning

International Journal of Academic and Industrial Research Innovations(IJAIRI)•Jun 11, 2026•11 citations

Explainable quantum-compressed machine learning for complex fluid flows

Xiao Xue, Maida Wang, Mingyang Gao +2 more

Machine-learning surrogates of physical systems face a paradox: explainable models facing the challenge of expressivity to capture complex nonlinear flows, whereas expressive deep surrogates match high-fidelity simulations only through massive parameterisations that turn the learned dynamics into a black box. Here, we introduce quantum-compressed machine learning (QCML), which resolves this tension by compressing the latent propagator of a flow surrogate from $524{,}288$ trainable parameters to no more than $8$. This parameter reduction brings the learned dynamical law to the parameter scale of a physical constitutive relation rather than a black-box neural network, making the surrogate directly interpretable and controllable without sacrificing expressivity. The compression is realised by a structured quantum circuit whose unitary propagator constrains the latent spectrum to the unit circle exactly and by construction, replacing exponential error growth with linear accumulation over autoregressive rollouts. Classical regularisation only approximates this constraint: even a quantum-inspired classical baseline penalised towards unitarity collapses within one Lyapunov time on turbulent channel flow, whereas QCML remains stable over the full rollout. Shared phase and coupling angles parameterising the circuit correspond directly to modal frequencies and inter-mode interactions, giving the learned dynamics a physical interpretation in spectral space. On two patient-specific cardiovascular benchmarks, the structured QCML propagator matches the predictive accuracy of its classical counterpart on surface pressure spectra, pressure drop, and wall shear stress. These results establish QCML as a working component of scientific machine learning and a concrete contribution towards practical quantum advantage in real-world prediction.

physics.flu-dynCross-listedarXiv:2607.21688•Jul 24, 2026•1 citation

Theory and interpretability of Quantum Extreme Learning Machines: a Pauli transfer matrix approach

Markus Gross, Hans-Martin Rieser

Quantum reservoir computers (QRCs) have emerged as a promising approach to quantum machine learning, since they utilize the natural dynamics of quantum systems for data processing and are simple to train. Here, we consider $n$-qubit quantum extreme learning machines (QELMs) with initial-state encoding and continuous-time reservoir dynamics. Applying the Pauli transfer matrix (PTM) formalism reveals the encoding-generated features and their linear transformation by quantum channels, and thereby provides a nonlinear vector autoregression model as an interpretable classical representation of a QELM. We show that the nonlinear information processing capacity of a QELM factorizes into the ability of the encoding to represent a given target function and the decodability of the relevant Pauli features after the effective PTM. Comparing geometric and statistical decodability reveals features that are algebraically hidden yet recoverable through data correlations. We relate the decodability of Pauli features to the spreading of measurement operators under Heisenberg evolution (Krylov growth). Focusing on the unitary channel, we discuss how structured Hamiltonians, when paired with certain observables, can impair model performance, as reflected in a low rank of the readout and the effective PTMs. We trace this effect to Hamiltonian symmetries and derive asymptotic rank estimates for symmetry-resolved observable families. Our decodability analysis disentangles the contributions of encoding, reservoir dynamics, and measurement observables, and thus provides interpretable diagnostics that can inform the design of QELMs and reduce reliance on exhaustive hyperparameter search. As a specific application, we consider forecasting nonlinear dynamical systems and show that the readout of a trained QELM closely matches the expression of the underlying flow map at leading orders.

quant-pharXiv:2602.18377•Physical Review A•Feb 23, 2026•2 citations
AI for Science aims not only to predict complex physical systems from data, but also to extract mathematical structure and physically testable representations from learned models. Here, a train--prune--readout--rewrite workflow is developed that separates physical-domain grounding from three increasingly stringent analysis claims: algebraically equivalent readout of a trained predictor, compact teacher-faithful symbolic rewriting on the sampled physical domain, and transformation-based tests of learned internal representations. The workflow is implemented with complex-valued Kolmogorov--Arnold networks, whose explicit edge functions enable post-pruning analytic readout of the retained computation. In analytically controlled single-qubit tasks, rewriting recovered the quadratic structure of purity, whereas von Neumann entropy yielded only a domain-bounded symbolic surrogate; physics-aligned variable grouping preserved symbolic fidelity. For two-qubit entanglement-related tasks, shared learning exposed a common internal representation whose physical content was interrogated directly. Local-unitary transformations rejected a direct invariant-coordinate interpretation, while fixed-decoder transfer showed that the shared activation carries Pauli-correlation information in a transformation-consistent form. Task-related invariant spectral features were subsequently recovered through low-order nonlinear readouts. Separate predictive tests retained high accuracy for three-qubit classification and controlled ten-qubit purity regression with over one million complex inputs. These results establish an evidence-resolved framework for distinguishing physical grounding, readable computation, faithful symbolic compression and transformation-tested physical structure in constrained complex-valued scientific learning.
Machine-learning approaches to quantum state tomography can achieve high reconstruction fidelity, but the physical structure used by the trained model often remains implicit. Here we ask whether a sparsified Kolmogorov-Arnold Network (KAN) can be used not only as a regressor, but also as an inspectable reconstruction rule whose internal organization can be checked against known Pauli structure. We study a controlled three-qubit GHZ-family benchmark in which all 63 non-identity Pauli expectation values are used to reconstruct three GHZ-subspace variables: the population imbalance zzz, the real off-diagonal component ccc, and the imaginary off-diagonal component sss. Under finite-shot sampling and depolarizing noise, external ablation identifies the extended 12-channel GHZ-relevant Pauli set from the 63 measurements, with exact top-12 recovery across the tested shot counts and depolarizing-noise strengths. These support patterns remain stable across multi-seed random-initialization and noise-level analyses, and collapse under random-label controls. The dominant pruned input-hidden-output pathways organize Z-type population observables and X/Y off-diagonal observables in a pattern consistent with the analytic GHZ Pauli grouping, and sparse formula recovery recovers the canonical signed Pauli relations. The contribution of the KAN is therefore pathway-level structural interpretability within a neural reconstruction model, rather than superior sparse regression. Together with negative controls, these probes provide a consistency chain for auditing learned reconstruction rules against known physical structure.
This paper introduces a novel quantum machine learning framework that integrates adaptive learning capabilities with superalgebraically structured entangled layers. Leveraging the mathematical formalism of Lie superalgebras—specifically osp(1∣2) and sl(2∣1)—the proposed architecture encodes quantum states in graded Hilbert spaces and governs their evolution via symmetry-preserving supercharges. The entangled layers dynamically adapt to task-specific feedback through a hybrid quantum-classical training protocol that ensures algebraic consistency and quantum coherence. Symbolic computation and quantum simulation tools are employed to validate the fidelity and symmetry of entanglement evolution. The system is benchmarked on tasks such as symbolic reasoning, quantum phase classification, and structured regression, outperforming baseline models in interpretability, stability, and generalization. This research lays the foundation for scalable, physically grounded quantum AI systems capable of learning through algebraic alignment with fundamental symmetries. Keywords: quantum machine learning, adaptive learning, entanglement, superalgebra, osp(1∣2), graded Hilbert space, quantum AI, symmetry-preserving learning, Lie superalgebras, supercharges, hybrid quantum-classical architecture, interpretable QML, symbolic reasoning, quantum circuits
We introduce Quantum Orthogonal Separable Physics-Informed Neural Networks (QO-SPINNs), which replace selected dense hidden transformations in physics-informed PDE models with quantum-compatible orthogonal maps. A pyramidal reconfigurable-beamsplitter circuit realizes these maps with Θ(d)Θ(d)Θ(d) circuit depth. Under the stated signed-readout access model and fixed per-call failure probability, a width-ddd map has sequential access-cost upper bound O(dlog⁡d/ε2)\mathcal O(d\log d/ε^2)O(dlogd/ε2), which is asymptotically smaller than the Θ(d2)Θ(d^2)Θ(d2) cost of conventional direct dense multiplication when ε(d)−2log⁡d=o(d)ε(d)^{-2}\log d=o(d)ε(d)−2logd=o(d). Because unary state preparation discards magnitude, we retain each hidden-state norm classically and restore it after readout, yielding the exact map WhWhWh under ideal readout. We study this construction in two complementary settings: forward PDE regression and a norm-controlled residual model for representation geometry and uncertainty quantification. In the norm-controlled model, structural orthogonality yields explicit bi-Lipschitz bounds, while finite-readout analysis relates sufficient reconstruction precision to a prescribed physical-coordinate resolution through trained-model constants. This makes the access-cost condition assessable for trained checkpoints, although favorable fixed-resolution scaling across widths is not established. A post-hoc Bayesian-linear covariance provides uncertainty from the same frozen predictor, and cross-PDE transfer shows that strong distance-awareness can persist even when absolute coverage calibration fails. All Quantum Orthogonal (QO) layers are evaluated by classical simulation: the results therefore characterize a quantum-compatible structured neural primitive rather than demonstrating a practical quantum-hardware advantage.
Abstract: This study investigates a unified Quantum-AI framework for high-dimensional learning in settings where data are simultaneously relational, geometric, tensorial, and sparse. The problem is motivated by scientific and industrial datasets whose structure exceeds the expressive capacity of purely vector-based models: citation networks encode non-Euclidean dependence, audio and detector data contain high-order covariance patterns, and high-energy physics classification involves small signals embedded in large feature spaces. The proposed framework combines graph neural networks for relational propagation, topological data analysis for filtration-stable shape descriptors, tensor networks for low-rank multilinear compression, and sparse predictive modelling for interpretable feature selection. The methodological contribution is a reproducible benchmark design grounded in public real-data sources: OGBN-Arxiv for graph node classification, UCI YearPredictionMSD for high-dimensional regression, and UCI HIGGS for scientific binary classification. Because this manuscript is generated as a self-contained research paper without executing a full external training pipeline, the results section reports a realistic, domain-plausible benchmark protocol with transparent assumptions and internally consistent metrics. Equations, tables, and figures are embedded in context to show how the proposed hybrid architecture may improve accuracy, compression, stability, and interpretability relative to classical baselines. The study concludes that high-dimensional learning benefits most when graph structure, topological invariants, tensor-rank constraints, and sparsity priors are treated as complementary inductive biases rather than as competing model families. Keywords Quantum artificial intelligence; high-dimensional learning; graph neural networks; topological data analysis; tensor networks; sparse predictive modelling; persistent homology; OGBN-Arxiv; HIGGS; YearPredictionMSD; quantum kernels; variational circuits; scalable machine learning
Machine-learning surrogates of physical systems face a paradox: explainable models facing the challenge of expressivity to capture complex nonlinear flows, whereas expressive deep surrogates match high-fidelity simulations only through massive parameterisations that turn the learned dynamics into a black box. Here, we introduce quantum-compressed machine learning (QCML), which resolves this tension by compressing the latent propagator of a flow surrogate from 524,288524{,}288524,288 trainable parameters to no more than 888. This parameter reduction brings the learned dynamical law to the parameter scale of a physical constitutive relation rather than a black-box neural network, making the surrogate directly interpretable and controllable without sacrificing expressivity. The compression is realised by a structured quantum circuit whose unitary propagator constrains the latent spectrum to the unit circle exactly and by construction, replacing exponential error growth with linear accumulation over autoregressive rollouts. Classical regularisation only approximates this constraint: even a quantum-inspired classical baseline penalised towards unitarity collapses within one Lyapunov time on turbulent channel flow, whereas QCML remains stable over the full rollout. Shared phase and coupling angles parameterising the circuit correspond directly to modal frequencies and inter-mode interactions, giving the learned dynamics a physical interpretation in spectral space. On two patient-specific cardiovascular benchmarks, the structured QCML propagator matches the predictive accuracy of its classical counterpart on surface pressure spectra, pressure drop, and wall shear stress. These results establish QCML as a working component of scientific machine learning and a concrete contribution towards practical quantum advantage in real-world prediction.
Quantum reservoir computers (QRCs) have emerged as a promising approach to quantum machine learning, since they utilize the natural dynamics of quantum systems for data processing and are simple to train. Here, we consider nnn-qubit quantum extreme learning machines (QELMs) with initial-state encoding and continuous-time reservoir dynamics. Applying the Pauli transfer matrix (PTM) formalism reveals the encoding-generated features and their linear transformation by quantum channels, and thereby provides a nonlinear vector autoregression model as an interpretable classical representation of a QELM. We show that the nonlinear information processing capacity of a QELM factorizes into the ability of the encoding to represent a given target function and the decodability of the relevant Pauli features after the effective PTM. Comparing geometric and statistical decodability reveals features that are algebraically hidden yet recoverable through data correlations. We relate the decodability of Pauli features to the spreading of measurement operators under Heisenberg evolution (Krylov growth). Focusing on the unitary channel, we discuss how structured Hamiltonians, when paired with certain observables, can impair model performance, as reflected in a low rank of the readout and the effective PTMs. We trace this effect to Hamiltonian symmetries and derive asymptotic rank estimates for symmetry-resolved observable families. Our decodability analysis disentangles the contributions of encoding, reservoir dynamics, and measurement observables, and thus provides interpretable diagnostics that can inform the design of QELMs and reduce reliance on exhaustive hyperparameter search. As a specific application, we consider forecasting nonlinear dynamical systems and show that the readout of a trained QELM closely matches the expression of the underlying flow map at leading orders.