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:2609.28391

Quantum score matching with applications to learning thermal states

Yulong Dong, Jiaqi Leng

quant-phcs.LGSubmitted Sep 23, 2026• Announced Sep 24, 2026• 2 citations
arXivHTMLPDF

Abstract

Score matching has driven major advances in classical generative learning by enabling models to learn from data without evaluating intractable normalization constants, or partition functions. Yet, extending this principle to quantum learning requires rethinking its foundations, as quantum states are described by noncommuting density operators rather than scalar probabilities. The noncommutativity creates fundamental challenges not only in defining quantum scores, but also in developing a training framework with efficient circuit implementations and rigorous theoretical guarantees. In this work, we bridge this gap by establishing a general quantum score-matching framework with end-to-end theoretical guarantees. Applied to Gibbs-state learning, our approach avoids additional thermal-state preparation and achieves information-theoretically optimal sample complexity in the high-temperature regime for Hamiltonians with bounded locality and interaction degree. This positions score matching as a new route to state-of-the-art performance in learning quantum Gibbs states. Beyond these theoretical results, numerical simulations show that our method remains effective even when gradients are estimated inaccurately under limited measurement budgets. Experiments on IBM quantum hardware further demonstrate that quantum score matching is NISQ-friendly: without any error mitigation or correction, it reduces the relative Hamiltonian-parameter error from 64% to approximately 10%. Together, these results extend score matching into an experimentally realizable paradigm for quantum-state learning.

AI Summary

The paper asks whether score matching can learn quantum states directly from copies without evaluating partition functions or preparing the current model’s Gibbs state. It defines operator-valued scores through commutator derivations and logarithmic derivatives, focusing on the symmetric logarithmic derivative. A quantum Hyvärinen identity rewrites squared score discrepancy as an expectation of a model-dependent observable on the target state. A commutant criterion supplies sufficient conditions for global identifiability. For Gibbs models $H(\theta)=\sum_{j=1}^m\theta_jP_j$, randomized real-time evolution and parameter-shift circuits estimate training gradients at any finite inverse temperature $\beta$, with polynomial copy cost and maximum evolution time $O(\beta\log(\operatorname{poly}(n,\beta)/\eta))$ for gradient accuracy $\eta$. For bounded-locality, bounded-overlap Hamiltonians at sufficiently high temperature, the single-qubit frame $\{X_x,Z_x\}$ suffices. Lieb–Robinson bounds establish quasi-local gradient observables, enabling shared-copy measurements; Hessian bounds establish convergence of preconditioned projected gradient descent. The resulting coefficient-learning guarantee is $\|\widehat\theta-\theta^*\|_\infty\leq\varepsilon$ with probability $1-\delta$, using $O(\beta^{-2}\varepsilon^{-2}\log(2m/\delta))$ copies, matching minimax bounds in the stated regime. Eight-qubit simulations examine finite-shot training. A four-qubit, two-parameter IBM demonstration reaches approximately 10% relative parameter error at $\beta=0.2$ after 45 updates, using 8,192 shots per update without error mitigation. The sample-optimal measurement theorem does not bound measurement-synthesis cost. General low-temperature efficiency is not guaranteed, and copy requirements can grow exponentially with $\beta$. Applying the identities beyond Gibbs families requires separate score representations.

Little sign of AI-generated textStrong assumptions

Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.

Comments
57 pages, 5 figures, 3 tables, with an accompanying GitHub repository at https://github.com/dongsnaq/Quantum-Score-Matching

Related papers

Operator Score Matching for Learning Quantum Hamiltonians

Shreya Shukla, Abhijith Jayakumar, Andrey Y. Lokhov

Learning quantum Hamiltonians from low-temperature thermal state measurements is a fundamental problem in quantum physics. Scalability of existing methods is limited by the complexity of semidefinite optimization problems or partition function computation. Here, we develop a quantum analog of classical score matching method that exploits generalized notions of derivatives and integration by parts in operator algebras. We introduce a new \emph{Operator Score Matching} loss function that recovers Hamiltonian parameters by a simple gradient descent without the need to compute intractable normalization constants. Numerical experiments on the Gibbs states of the transverse-field Ising model, XXZ spin chain, Fermi-Hubbard model, and lattice $φ^4$ theory demonstrate efficient parameter recovery using finite-depth truncations of nested commutator expansions. Our results establish Operator Score Matching as a practical, partition-function-free framework for quantum Hamiltonian learning.

quant-pharXiv:2609.25246•Sep 23, 2026•2 citations

Quantum score matching with applications to learning thermal states — E8 Intelligence Research

Andrew Stewart Caldin

Score matching has driven major advances in classical generative learning by enabling models to learn from data without evaluating intractable normalization constants, or partition functions. Yet, extending this principle to quantum learning requires rethinking its foundations, as quantum states are described by noncommuting density operators rather than scalar probabilities. The noncommutativity creates fundamental challenges not only in defining quantum scores, but also in developing a training framework with efficient circuit implementations and rigorous theoretical guarantees. In this work, we bridge this gap by establishing a general quantum score-matching framework with end-to-end theoretical guarantees. Applied to Gibbs-state learning, our approach avoids additional thermal-state preparation and achieves information-theoretically optimal sample complexity in the high-temperature regime for Hamiltonians with bounded locality and interaction degree. This positions score matching a Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com

•Nov 11, 2026•0 citations

Supervised Learning with Quantum Hamiltonian Embeddings

Riccardo Molteni, Minjun Jeon, Paolo Braccia +8 more

Machine learning is widely regarded as a promising application in which quantum computing could offer a computational advantage. However, general quantum machine learning models with provable guarantees on performance, trainability, and efficient inference remain limited. In this work, we introduce quantum Hamiltonian embeddings, a supervised learning framework in which a classical training dataset is encoded into a single effective, parameterized Hamiltonian. Training optimizes the parameters of this data-dependent Hamiltonian, whose approximate ground state defines the classification rule. Once the ground state has been approximated using a quantum algorithm, inference can be performed either classically using a classical representation of the learned state, or with a quantum representation of the state. We establish efficient methods for gradient estimation and prove favorable generalization guarantees. Furthermore, under standard cryptographic assumptions, we show that Hamiltonian embeddings can realize hypothesis functions beyond the reach of efficient classical algorithms. We then demonstrate the framework at scale on IBM Heron processors using a standard benchmark dataset. Together, these results establish Hamiltonian embeddings as a new framework for quantum machine learning, combining theoretical guarantees with a path for experimental investigations at scale on quantum hardware.

quant-pharXiv:2610.26833•Oct 20, 2026•0 citations

Quantum Gibbs Sampling in Infinite Dimensions: Generation, Mixing Times and Circuit Implementation

Simon Becker, Cambyse Rouzé, Robert Salzmann

We develop a rigorous and implementable framework for Gibbs sampling of infinite-dimensional quantum systems governed by unbounded Hamiltonians. Extending dissipative Gibbs samplers beyond finite dimensions raises fundamental obstacles, including ill-defined generators, the absence of spectral gaps on natural Banach spaces, and tensions between implementability and convergence guarantees. We overcome these issues by constructing KMS-symmetric quantum Markov semigroups on separable Hilbert spaces that are both well-posed and efficiently implementable on qubit hardware. Our generation theory is based on the abstract framework of Dirichlet forms, adapted here to the case of algebras of bounded operators over separable Hilbert spaces. Leveraging the spectral properties of our self-adjoint generators, we establish quantitative convergence results in trace distance, including regimes of fast thermalization. In contrast, we also identify Hamiltonians for which a naive choice of generators guaranteeing implementability generally comes at the cost of losing uniform exponential convergence guarantees for the associated evolutions, thereby establishing a strong trade-off between implementability and convergence. Our framework applies to a wide class of models - including Schrödinger operators, Gaussian systems, and Bose-Hubbard Hamiltonians - and provides a unified approach linking rigorous infinite-dimensional analysis with algorithmic Gibbs state preparation.

quant-pharXiv:2604.01192•Apr 2, 2026•6 citations

Learning quantum Gibbs states locally and efficiently

Chi-Fang Chen, Anurag Anshu, Quynh T. Nguyen

Learning the Hamiltonian underlying a quantum many-body system in thermal equilibrium is a fundamental task in quantum learning theory and experimental sciences. To learn the Gibbs state of local Hamiltonians at any constant inverse temperature, the state-of-the-art provable algorithms fall short of the optimal sample and computational complexity, in sharp contrast with the locality and simplicity in the classical cases. In this work, we present a learning algorithm that learns each local term of a n-qubit Hamiltonian on any bounded-degree graph to a constant additive error with the optimal sample complexity $\mathcal{O}(\log n)$. The protocol uses parallelizable local quantum measurements that act within bounded neighborhoods of the graph and near-linear-time classical post-processing. We also give a learning algorithm for lattice Hamiltonians with near-optimal scaling on the learning precision and the inverse temperature. At the heart of our algorithm is the interplay between locality, the Kubo-MartinSchwinger condition, and the operator Fourier transform at arbitrary temperatures.

arXiv:2504.02706•Annual Symposium on Foundations of Computer Science•Dec 14, 2025•26 citations

Near-Optimal Learning of Quantum Many-Body Systems from Gibbs and Thermal Metastable States via the Stationarity Test

Tai Hsuan Yang

We study the problem of learning a finite-range quantum Hamiltonian on an $n$-qubit $D$-dimensional lattice from copies of its Gibbs states at a known inverse temperature $β$ or its thermal metastable states. Existing efficient algorithms at arbitrary temperature have sample guarantees with an $e^{\poly(β)}$ dependence, leaving open whether the exponent can be made linear in $β$. We achieve this improvement: for $β\ge1$, our algorithm learns all Hamiltonian coefficients to $\ell_\infty$-error $ε$ with probability at least $1-δ$ using $\tO\bigl(e^{O(β)}ε^{-2}\log(n/δ)\bigr)$ copies and total time $\tO\bigl(n e^{O(β)}ε^{-2}\log(n/δ)\bigr)$. The measurements act on individual copies and are supported on balls of radius $O\bigl(β^2+β\log(1/ε)\bigr)$. Our sample complexity achieves the optimal exponential order in $β$, with optimal $ε^{-2}$ dependence up to logarithmic factors. Building on the stationarity-test framework \citep{Bergamaschi2026Stationarity}, we prove a unified Fisher-information coercivity bound showing that errors in the candidate Hamiltonian produce a detectable signal with only an $e^{-O(β)}$ loss in sensitivity. The key is a smooth frequency filter whose bandwidth remains independent of temperature, together with sharper estimates relating this signal to the Hamiltonian coefficients. For thermal metastable states, the guarantee holds under local drift $ζ\le e^{-Cβ^{2D+1}}$, for a geometry-dependent constant $C>0$, and above an accuracy floor $\tO(e^{O(β)}ζ)$. The argument applies to both Gibbs and thermal metastable states and, combined with local measurements and projected stochastic gradient descent (SGD), yields both learning guarantees.

quant-pharXiv:2611.01520•Nov 3, 2026•0 citations
Score matching has driven major advances in classical generative learning by enabling models to learn from data without evaluating intractable normalization constants, or partition functions. Yet, extending this principle to quantum learning requires rethinking its foundations, as quantum states are described by noncommuting density operators rather than scalar probabilities. The noncommutativity creates fundamental challenges not only in defining quantum scores, but also in developing a training framework with efficient circuit implementations and rigorous theoretical guarantees. In this work, we bridge this gap by establishing a general quantum score-matching framework with end-to-end theoretical guarantees. Applied to Gibbs-state learning, our approach avoids additional thermal-state preparation and achieves information-theoretically optimal sample complexity in the high-temperature regime for Hamiltonians with bounded locality and interaction degree. This positions score matching as a new route to state-of-the-art performance in learning quantum Gibbs states.
Beyond these theoretical results, numerical simulations show that our method remains effective even when gradients are estimated inaccurately under limited measurement budgets. Experiments on IBM quantum hardware further demonstrate that quantum score matching is NISQ-friendly: without any error mitigation or correction, it reduces the relative Hamiltonian-parameter error from 64% to approximately 10%. Together, these results extend score matching into an experimentally realizable paradigm for quantum-state learning.
The paper asks whether score matching can learn quantum states directly from copies without evaluating partition functions or preparing the current model’s Gibbs state. It defines operator-valued scores through commutator derivations and logarithmic derivatives, focusing on the symmetric logarithmic derivative. A quantum Hyvärinen identity rewrites squared score discrepancy as an expectation of a model-dependent observable on the target state. A commutant criterion supplies sufficient conditions for global identifiability.
For Gibbs models H(θ)=∑j=1mθjPjH(\theta)=\sum_{j=1}^m\theta_jP_jH(θ)=∑j=1m​θj​Pj​, randomized real-time evolution and parameter-shift circuits estimate training gradients at any finite inverse temperature β\betaβ, with polynomial copy cost and maximum evolution time O(βlog⁡(poly⁡(n,β)/η))O(\beta\log(\operatorname{poly}(n,\beta)/\eta))O(βlog(poly(n,β)/η)) for gradient accuracy η\etaη. For bounded-locality, bounded-overlap Hamiltonians at sufficiently high temperature, the single-qubit frame {Xx,Zx}\{X_x,Z_x\}{Xx​,Zx​} suffices. Lieb–Robinson bounds establish quasi-local gradient observables, enabling shared-copy measurements; Hessian bounds establish convergence of preconditioned projected gradient descent. The resulting coefficient-learning guarantee is ∥θ^−θ∗∥∞≤ε\|\widehat\theta-\theta^*\|_\infty\leq\varepsilon∥θ−θ∗∥∞​≤ε with probability 1−δ1-\delta1−δ, using O(β−2ε−2log⁡(2m/δ))O(\beta^{-2}\varepsilon^{-2}\log(2m/\delta))O(β−2ε−2log(2m/δ)) copies, matching minimax bounds in the stated regime.
Eight-qubit simulations examine finite-shot training. A four-qubit, two-parameter IBM demonstration reaches approximately 10% relative parameter error at β=0.2\beta=0.2β=0.2 after 45 updates, using 8,192 shots per update without error mitigation. The sample-optimal measurement theorem does not bound measurement-synthesis cost. General low-temperature efficiency is not guaranteed, and copy requirements can grow exponentially with β\betaβ. Applying the identities beyond Gibbs families requires separate score representations.
Learning quantum Hamiltonians from low-temperature thermal state measurements is a fundamental problem in quantum physics. Scalability of existing methods is limited by the complexity of semidefinite optimization problems or partition function computation. Here, we develop a quantum analog of classical score matching method that exploits generalized notions of derivatives and integration by parts in operator algebras. We introduce a new \emph{Operator Score Matching} loss function that recovers Hamiltonian parameters by a simple gradient descent without the need to compute intractable normalization constants. Numerical experiments on the Gibbs states of the transverse-field Ising model, XXZ spin chain, Fermi-Hubbard model, and lattice φ4φ^4φ4 theory demonstrate efficient parameter recovery using finite-depth truncations of nested commutator expansions. Our results establish Operator Score Matching as a practical, partition-function-free framework for quantum Hamiltonian learning.
Score matching has driven major advances in classical generative learning by enabling models to learn from data without evaluating intractable normalization constants, or partition functions. Yet, extending this principle to quantum learning requires rethinking its foundations, as quantum states are described by noncommuting density operators rather than scalar probabilities. The noncommutativity creates fundamental challenges not only in defining quantum scores, but also in developing a training framework with efficient circuit implementations and rigorous theoretical guarantees. In this work, we bridge this gap by establishing a general quantum score-matching framework with end-to-end theoretical guarantees. Applied to Gibbs-state learning, our approach avoids additional thermal-state preparation and achieves information-theoretically optimal sample complexity in the high-temperature regime for Hamiltonians with bounded locality and interaction degree. This positions score matching a Author: Andrew Stewart Caldin, Independent Researcher, UK. Part of the E8 Intelligence Research series. Platform: e8intelligence.com
Machine learning is widely regarded as a promising application in which quantum computing could offer a computational advantage. However, general quantum machine learning models with provable guarantees on performance, trainability, and efficient inference remain limited. In this work, we introduce quantum Hamiltonian embeddings, a supervised learning framework in which a classical training dataset is encoded into a single effective, parameterized Hamiltonian. Training optimizes the parameters of this data-dependent Hamiltonian, whose approximate ground state defines the classification rule. Once the ground state has been approximated using a quantum algorithm, inference can be performed either classically using a classical representation of the learned state, or with a quantum representation of the state. We establish efficient methods for gradient estimation and prove favorable generalization guarantees. Furthermore, under standard cryptographic assumptions, we show that Hamiltonian embeddings can realize hypothesis functions beyond the reach of efficient classical algorithms. We then demonstrate the framework at scale on IBM Heron processors using a standard benchmark dataset. Together, these results establish Hamiltonian embeddings as a new framework for quantum machine learning, combining theoretical guarantees with a path for experimental investigations at scale on quantum hardware.
We develop a rigorous and implementable framework for Gibbs sampling of infinite-dimensional quantum systems governed by unbounded Hamiltonians. Extending dissipative Gibbs samplers beyond finite dimensions raises fundamental obstacles, including ill-defined generators, the absence of spectral gaps on natural Banach spaces, and tensions between implementability and convergence guarantees. We overcome these issues by constructing KMS-symmetric quantum Markov semigroups on separable Hilbert spaces that are both well-posed and efficiently implementable on qubit hardware. Our generation theory is based on the abstract framework of Dirichlet forms, adapted here to the case of algebras of bounded operators over separable Hilbert spaces. Leveraging the spectral properties of our self-adjoint generators, we establish quantitative convergence results in trace distance, including regimes of fast thermalization. In contrast, we also identify Hamiltonians for which a naive choice of generators guaranteeing implementability generally comes at the cost of losing uniform exponential convergence guarantees for the associated evolutions, thereby establishing a strong trade-off between implementability and convergence. Our framework applies to a wide class of models - including Schrödinger operators, Gaussian systems, and Bose-Hubbard Hamiltonians - and provides a unified approach linking rigorous infinite-dimensional analysis with algorithmic Gibbs state preparation.
Learning the Hamiltonian underlying a quantum many-body system in thermal equilibrium is a fundamental task in quantum learning theory and experimental sciences. To learn the Gibbs state of local Hamiltonians at any constant inverse temperature, the state-of-the-art provable algorithms fall short of the optimal sample and computational complexity, in sharp contrast with the locality and simplicity in the classical cases. In this work, we present a learning algorithm that learns each local term of a n-qubit Hamiltonian on any bounded-degree graph to a constant additive error with the optimal sample complexity O(log⁡n)\mathcal{O}(\log n)O(logn). The protocol uses parallelizable local quantum measurements that act within bounded neighborhoods of the graph and near-linear-time classical post-processing. We also give a learning algorithm for lattice Hamiltonians with near-optimal scaling on the learning precision and the inverse temperature. At the heart of our algorithm is the interplay between locality, the Kubo-MartinSchwinger condition, and the operator Fourier transform at arbitrary temperatures.
We study the problem of learning a finite-range quantum Hamiltonian on an nnn-qubit DDD-dimensional lattice from copies of its Gibbs states at a known inverse temperature βββ or its thermal metastable states. Existing efficient algorithms at arbitrary temperature have sample guarantees with an $e^{\poly(β)}$ dependence, leaving open whether the exponent can be made linear in βββ. We achieve this improvement: for β≥1β\ge1β≥1, our algorithm learns all Hamiltonian coefficients to ℓ∞\ell_\inftyℓ∞​-error εεε with probability at least 1−δ1-δ1−δ using $\tO\bigl(e^{O(β)}ε^{-2}\log(n/δ)\bigr)$ copies and total time $\tO\bigl(n e^{O(β)}ε^{-2}\log(n/δ)\bigr)$. The measurements act on individual copies and are supported on balls of radius O(β2+βlog⁡(1/ε))O\bigl(β^2+β\log(1/ε)\bigr)O(β2+βlog(1/ε)). Our sample complexity achieves the optimal exponential order in βββ, with optimal ε−2ε^{-2}ε−2 dependence up to logarithmic factors. Building on the stationarity-test framework \citep{Bergamaschi2026Stationarity}, we prove a unified Fisher-information coercivity bound showing that errors in the candidate Hamiltonian produce a detectable signal with only an e−O(β)e^{-O(β)}e−O(β) loss in sensitivity. The key is a smooth frequency filter whose bandwidth remains independent of temperature, together with sharper estimates relating this signal to the Hamiltonian coefficients. For thermal metastable states, the guarantee holds under local drift ζ≤e−Cβ2D+1ζ\le e^{-Cβ^{2D+1}}ζ≤e−Cβ2D+1, for a geometry-dependent constant C>0C>0C>0, and above an accuracy floor $\tO(e^{O(β)}ζ)$. The argument applies to both Gibbs and thermal metastable states and, combined with local measurements and projected stochastic gradient descent (SGD), yields both learning guarantees.