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:2610.40393

Quantum Computing through Computational Imaging: Representations, Measurement, and Inference

George Kourousias, Sergio Carrato, Roberto Pugliese, Panagiotis Dimitrakis, Alberto Carini

quant-phSubmitted Oct 29, 2026• Announced Oct 30, 2026• 0 citations
arXivHTMLPDF

Abstract

An algebraic correspondence between an imaging model and a quantum state does not specify a computational method: input access, measurement and decoding determine what can be returned. This exposition develops four distinctions for computational-imaging and engineering readers. It separates amplitude encodings of input values from encodings of complete candidate labelings, explaining why amplitude storage does not provide full-array readout. Complex-field interference motivates unknown-phase inference: complementary measurements resolve ambiguities that local Fisher information alone does not detect. Reshaping joint amplitudes into a matrix connects singular-value decomposition to pure-state entanglement, while distinguishing Schmidt rank from conditions for efficient classical simulation. In a four-pixel segmentation, an objective becomes a cost operator and a sampling circuit. One layer of the quantum approximate optimization algorithm (QAOA) lowers mean energy from $1.675$ to $0.686$ relative to uniform sampling, yet makes a nonoptimal labeling more probable than the optimum 0.484 versus 0.365. Archived IBM and IQM measurements illustrate a separate six-variable local boundary decision. Low-rank state reconstruction and selected quantum algorithms connect these distinctions to an end-to-end resource account. The derivations and reproducible numerical references assume complex linear algebra and probability, without prior quantum mechanics. The accompanying Seeing Amplitudes browser laboratory supports interactive exploration; the article remains self-contained and connects each representation to the estimate or decision it permits and the resources that output requires.

AI Summary

This tutorial asks how computational-imaging representations translate into quantum computations with specified inputs, measurements and decoded outputs. It develops four distinctions: amplitude encoding versus encoding complete candidate labelings; global identifiability versus local precision; pure-state entanglement versus classical simulation difficulty; and expected objective versus modal or best-sampled decisions. The analysis uses circuit derivations, statistical estimation and reproducible small-instance simulations. For an unknown single-qubit phase, complementary $X/Y$ measurements resolve the ambiguity remaining in an $X$ record alone. With $N$ shots per setting, local Fisher information is $2N$ at regular phases, giving a variance benchmark $1/(2N)$. The simple $\operatorname{atan2}$ estimator has asymptotic variance $(\sin^4\phi+\cos^4\phi)/N$, between one and two times that benchmark. Reshaping joint amplitudes connects SVD to Schmidt decomposition, while tensor-network and stabilizer examples explain why entanglement alone does not determine simulation cost. A four-pixel segmentation is mapped through QUBO and Ising forms to depth-one QAOA. Grid-selected angles lower expected energy from $1.675$ to $0.686$, but the nonoptimal labeling has probability $0.484$, exceeding the optimum’s $0.365$. Separate archived six-variable IBM/IQM measurements illustrate decoding a local boundary event rather than a full assignment. An accompanying browser laboratory supports exploration. The examples establish no quantum advantage: the segmentation admits exact polynomial-time minimum-cut solution, parameter selection uses exact simulated expectations on a restricted grid, and the hardware data concern one fixed synthetic task. Preparation, access, acquisition, decoding and classical baselines must be included in resource comparisons; agreement with a model optimum is not validation against physical truth.

Little sign of AI-generated textAI use disclosed
The LLMs GPT-6.1 Sol and Claude Opus~5.5 supported selected explanatory passages on phase inference, matrix factorization and imaging decisions, as well as revisions, figure refinement, code and consistency checks

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

Comments
v1: Supplementary code, data and web companion: https://doi.org/10.34965/enodo.i61162

Related papers

Quantum Computing Preliminaries: The Mathematical Language of Quantum Computing

P. K. Majumdar

Quantum computing uses a mathematical language that can seem unfamiliar to readers from programming, engineering and other technical backgrounds. Qubits are represented by vectors, quantum gates by matrices, measurement outcomes by eigenvalues, and probability amplitudes by complex numbers. Quantum Computing Preliminaries develops this language carefully, connecting each mathematical idea with its computational and physical meaning. The book begins with complex vector spaces and explains why quantum states require more than classical probabilities. It introduces qubits, superposition, normalisation, basis vectors and Dirac bra–ket notation, supported by worked calculations and geometric interpretation using the Bloch sphere. The discussion then turns to inner products and the geometry of quantum states. Readers learn how to calculate norms, test orthogonality, determine state overlap and apply the Born rule. Particular attention is given to complex conjugation and the distinction between the ordinary dot product and the Hermitian inner product used in quantum mechanics.The treatment of matrix operators establishes the roles of linear, unitary and Hermitian operators. It explains matrix multiplication, non-commutativity, eigenvalues, eigenvectors, reversibility and norm preservation. The Pauli and Hadamard gates provide concrete examples that link matrix algebra to rotations, the Bloch sphere, quantum-state evolution, and measurement. Outer products and projection operators are developed as tools for constructing operators, selecting state components and describing ideal projective measurements. The book examines idempotence, completeness, repeated measurement and the distinction between the ket–bra outer product and the tensor product used to form composite quantum systems. The final chapter explores changes of basis, operator representations and diagonalisation. It shows how the same physical state or operation can have different numerical representations in the computational and Hadamard bases. Readers derive the similarity transformation, use outer-product substitution, diagonalise Hermitian operators and interpret the spectral decomposition geometrically. Examples involving the Pauli-X, Pauli-Y, Pauli-Z and Hadamard operators demonstrate how basis choice exposes phase, population and measurement structure without changing the underlying physics. Throughout the book, Python and NumPy examples translate the mathematics into executable calculations. The programs demonstrate complex-vector operations, inner products, eigendecomposition, basis transformation and numerical verification while explaining the importance of using operations suited to complex Hilbert spaces. Each chapter includes conceptual, mathematical, computational and challenge exercises. Detailed solutions provide step-by-step derivations, program output and physical interpretation, allowing the book to support both guided study and independent practice. Quantum Computing Preliminaries is intended for students, software developers, engineers and technically minded readers preparing to study quantum circuits, algorithms, simulation, measurement, entanglement or error correction. It assumes familiarity with elementary algebra and programming but does not require prior training in quantum mechanics. Its purpose is to provide a clear mathematical foundation for understanding how quantum information is represented, transformed and measured.

•Sep 4, 2026•0 citations

Schmidt Decomposition-Based Methods for Efficient Quantum Image Encoding

Ana-Maria Pangeva, Yassine Ferhi, Alexander Geng +3 more

In quantum image processing, a fundamental step is encoding classical image data into quantum states. This can be achieved using methods such as Flexible Representation of Quantum Images (FRQI), Quantum Probability Image Encoding (QPIE), and Novel Enhanced Quantum Representation (NEQR). However, on real quantum hardware, these encodings can quickly lead to circuits with many gates, large circuit depth, and high qubit usage, which is a problem for Noisy Intermediate-Scale Quantum (NISQ) devices. In this work, we investigate whether low-rank state approximation, formulated via Schmidt decomposition, can help reduce this complexity. The method keeps only the most significant parts of a quantum state's entanglement structure, making state preparation more efficient while preserving most of the image information. We compare the three encoding techniques in their original form and with low-rank approximation, evaluating metrics such as circuit depth, CNOT count, MSE, and visual quality of reconstructed images. The results reveal meaningful trade-offs between accuracy and resource efficiency, with the FRQI model achieving a 97 percent reduction in circuit depth while maintaining a near-perfect reconstruction (MSE of about 0.27). This demonstrates the potential of low-rank techniques for advancing practical quantum image processing on near-term hardware.

cs.CVCross-listedarXiv:2606.10874•Jun 10, 2026•0 citations

Quantum Linear Algebra for Machine Learning

Hamid D. Ismail

This chapter examines quantum linear algebra as a foundational framework for quantum machine learning. Because many core learning tasks, including regression, dimensionality reduction, kernel evaluation, recommendation systems, and representation learning, depend on linear algebraic operations, this chapter explores how these operations can be reformulated in the quantum setting. Emphasis is placed on the shift from explicit numerical computation to coherent manipulation of quantum states and measurement-based access to information. This chapter introduces quantum inner product estimation as a primitive for similarity and kernel evaluation, highlighting the swap test and related overlap-estimation methods. It then develops quantum matrix inversion and the HHL algorithm, emphasizing the preparation of solution states rather than explicit solution vectors and discussing the importance of conditioning and regularization. Quantum singular value decomposition is presented through Hermitian embeddings, phase estimation, block encoding, and quantum singular value transformation, providing a unified framework for spectral filtering and low-rank approximation. This chapter concludes with quantum principal component analysis, in which covariance structure is represented by a density matrix and accessed through quantum phase estimation and density matrix exponentiation. Throughout, practical demonstrations in Qiskit and PennyLane illustrate the algorithms, while careful attention is given to the limitations imposed by state preparation, measurement, noise, and classical data loading.

CRC Press eBooks•Nov 4, 2026•0 citations

Introduction to quantum computing

Pawan Kumar Verma, Mohit Kumar, Saurabh Singhal +1 more

This chapter introduces quantum computing as a new computational paradigm built on superposition, entanglement, interference, and measurement and establishes its relevance to quantum searching and optimization. It first defines the core information primitives—qubits, quantum gates, and quantum circuits—and explains how they enable nonclassical state manipulation compared with deterministic bit-based processing. This chapter then surveys three major models of quantum computation: the gate-based circuit model for algorithmic execution via unitary gate sequences, the measurement-based model using pre-entangled resource states with computation driven by adaptive measurements, and the adiabatic model that encodes solutions in the ground state of a gradually evolved system, aligning naturally with annealing-style optimization. A structured comparison between classical and quantum computing highlights differences in data representation, intrinsic parallelism, scalability constraints, and error behavior and situates quantum supremacy as a task-specific benchmark rather than a universal replacement for classical machines. This chapter connects these foundations to application-critical algorithms, emphasizing Grover’s search and Shor’s factoring as canonical demonstrations of computational advantage. Finally, it motivates the book’s focus by mapping quantum search and optimization to practical domains including database querying, cryptography and post-quantum security, financial risk and fraud analytics, supply-chain logistics, scientific simulation, and emerging workflows such as quantum-assisted drug discovery and routing problems, while noting current hardware limitations and the need for robust error control.

CRC Press eBooks•Nov 6, 2026•0 citations

Optimal entanglement weights for heterodyne uncertainty with finite quantum memory

Zixuan He

A finite quantum memory is specified by more than its entanglement entropy. This paper determines how to shape its Schmidt weights and allocate optical photons to minimize the quantum conditional uncertainty of a prescribed heterodyne measurement. The resulting resource laws let researchers compare a uniformly weighted entangled source with a shaped source, and assess when sharing one memory across optical modes is more efficient than independently preparing mode–memory pairs. Complete preparation model and sharp leading resource law. Let \(V_{m,D}(E)\) be the minimum of \(H_\mu(Z|B)=h_\mu(Z)-\chi(Z:B)\) over all normal, possibly mixed optical–memory states with total mean photon number at most \(E\) and total accessible memory dimension at most \(D\). Here \(Z\) is the outcome of an \(m\)-mode heterodyne measurement, \(B\) is the retained quantum memory, \(\chi\) is Holevo information, and \(d\mu=\pi^{-m}d^{2m}z\). Accessible flags count toward the same memory budget. For every fixed \(m\) and every joint limit \(D,E\to\infty\), \[V_{m,D}(E)\sim\frac{\Lambda_m(E/T_D)}{T_D},\qquad T_D=(m!D)^{1/m}.\] The function \(\Lambda_m\) is an explicitly characterized constrained radial ground-state energy. Its Kummer and Bessel profiles give legal attaining preparations. The converse retains arbitrary coherent inputs and their photon tails, rather than optimizing only a diagonal ansatz. Why shaping changes the resource exponent. At the common photon ceiling \(E=T_D\), optimally shaped sources have uncertainty of order \(D^{-1/m}\), whereas final flat Schmidt spectra have optimal order \(D^{-1/(2m)}\), even after optimizing the occupied optical subspace. Flat weights create a sharp support boundary; tapering the weights replaces its boundary-layer cost by a smaller gradient cost. To reach a small target uncertainty \(\varepsilon\), this changes the leading memory requirement from \(2m\log_2(1/\varepsilon)\) to \(m\log_2(1/\varepsilon)\), and the attainable photon scaling from quadratic to linear in \(1/\varepsilon\). This improvement is compatible with nearly maximal entanglement entropy. For the explicit canonical saturated construction, \[S(\rho_{B,D})=\ln D-c_m+o(1),\qquad \|\rho_{B,D}-I_D/D\|_1\longrightarrow L_m>0.\] The full trace-norm limits are approximately \(L_1=0.8771\) and \(L_2=1.1170\); the half-trace-distance convention halves these numbers. Thus a vanishing relative entropy deficit does not make the optimized spectrum uniform. Collective preparation at fully counted resources. For two modes in the energy-sufficient asymptotic regime, sharing quantum memory reduces the leading conditional entropy by about \(10.24\%\) relative to the optimized independent mode–memory preparation class. That comparator includes local state optimization, integer memory and energy allocations, mixtures and charged accessible flags; it is not the class of all separable joint states. A separate finite construction uses seven coherent optical inputs, three retained qubits and \(0.21\) input photons. Under the specified loss, phase and stochastic memory-fault contract, its retained photon budget is at most \(0.200781\), and its conditional entropy is strictly lower than that of the same independent comparator by more than \(0.00547535\) nats. The integrated article gives the preparation circuit, one-sided numerical bounds and complete same-source error accounting. Interpretation. The objective uses quantum side information. A classical measurement of the memory generally has accessible information below \(\chi(Z:B)\), so the conditional-entropy advantages do not by themselves assert the same improvement in estimation precision or classical readout performance. The finite circuit is a rigorously specified source model, not a qualification of an implemented laboratory apparatus. The single PDF contains the article, five figures, full proofs and references.

•Oct 25, 2026•0 citations

Quantum Source Completions and Memory Environment Tradeoffs

Jeremy Rodgers

How much internal quantum memory is needed to reproduce a stationary stochastic process, and how does reducing that memory change the environmental resources required when output records are retained? This paper establishes exact quantum-source constructions and resource bounds under explicitly specified readout and retained-output conditions. A general completion theorem identifies the minimum Hilbert-space dimension for stationary triple data in the orthogonal-current-symbol source class: it equals the sum of the complex positive-semidefinite ranks of the middle-symbol slices. An attaining construction reproduces the entire canonical stationary order-two process. For the delayed trine process, the minimum classical hidden Markov model requires nine states at every non-independent parameter value. Quantum instruments reproduce the same complete output law using four-dimensional memory throughout the parameter range, with four dimensions proved necessary in a certified low-noise regime and three dimensions both sufficient and necessary on an explicit high-noise interval. The paper then separates reusable memory dimension from environmental dilation cost. On the certified high-noise interval, the exact one-step Pareto frontier consists of a three-dimensional source with recorded-channel Choi rank nine and a four-dimensional source with rank three. Here, Choi rank measures the minimum initially pure environmental remainder dimension for the channel retaining both the emitted classical symbol and final quantum memory. A separate mixed-support qutrit construction at noise parameter 0.99 improves complete-history costs and gives upper bounds close to universal lower bounds. These results show why optimizing a single step need not optimize a retained multi-step history. The analysis also distinguishes complete channels, fixed-input states and archive-only outputs, and demonstrates how access to a retained reference can distinguish generators with identical passive output statistics. Full-word proofs, algebraic complete-positivity certificates, a technical companion and reproducibility materials accompany the results. The exact intermediate memory boundary and complete-history optimum remain open. All results use standard quantum probabilities and apply to the stated operational resource models; they do not establish total hardware, energy or thermodynamic minima.

•Oct 28, 2026•0 citations
An algebraic correspondence between an imaging model and a quantum state does not specify a computational method: input access, measurement and decoding determine what can be returned. This exposition develops four distinctions for computational-imaging and engineering readers. It separates amplitude encodings of input values from encodings of complete candidate labelings, explaining why amplitude storage does not provide full-array readout. Complex-field interference motivates unknown-phase inference: complementary measurements resolve ambiguities that local Fisher information alone does not detect. Reshaping joint amplitudes into a matrix connects singular-value decomposition to pure-state entanglement, while distinguishing Schmidt rank from conditions for efficient classical simulation. In a four-pixel segmentation, an objective becomes a cost operator and a sampling circuit. One layer of the quantum approximate optimization algorithm (QAOA) lowers mean energy from 1.6751.6751.675 to 0.6860.6860.686 relative to uniform sampling, yet makes a nonoptimal labeling more probable than the optimum 0.484 versus 0.365. Archived IBM and IQM measurements illustrate a separate six-variable local boundary decision. Low-rank state reconstruction and selected quantum algorithms connect these distinctions to an end-to-end resource account. The derivations and reproducible numerical references assume complex linear algebra and probability, without prior quantum mechanics. The accompanying Seeing Amplitudes browser laboratory supports interactive exploration; the article remains self-contained and connects each representation to the estimate or decision it permits and the resources that output requires.
This tutorial asks how computational-imaging representations translate into quantum computations with specified inputs, measurements and decoded outputs. It develops four distinctions: amplitude encoding versus encoding complete candidate labelings; global identifiability versus local precision; pure-state entanglement versus classical simulation difficulty; and expected objective versus modal or best-sampled decisions.
The analysis uses circuit derivations, statistical estimation and reproducible small-instance simulations. For an unknown single-qubit phase, complementary X/YX/YX/Y measurements resolve the ambiguity remaining in an XXX record alone. With NNN shots per setting, local Fisher information is 2N2N2N at regular phases, giving a variance benchmark 1/(2N)1/(2N)1/(2N). The simple atan2⁡\operatorname{atan2}atan2 estimator has asymptotic variance (sin⁡4ϕ+cos⁡4ϕ)/N(\sin^4\phi+\cos^4\phi)/N(sin4ϕ+cos4ϕ)/N, between one and two times that benchmark. Reshaping joint amplitudes connects SVD to Schmidt decomposition, while tensor-network and stabilizer examples explain why entanglement alone does not determine simulation cost.
A four-pixel segmentation is mapped through QUBO and Ising forms to depth-one QAOA. Grid-selected angles lower expected energy from 1.6751.6751.675 to 0.6860.6860.686, but the nonoptimal labeling has probability 0.4840.4840.484, exceeding the optimum’s 0.3650.3650.365. Separate archived six-variable IBM/IQM measurements illustrate decoding a local boundary event rather than a full assignment. An accompanying browser laboratory supports exploration.
The examples establish no quantum advantage: the segmentation admits exact polynomial-time minimum-cut solution, parameter selection uses exact simulated expectations on a restricted grid, and the hardware data concern one fixed synthetic task. Preparation, access, acquisition, decoding and classical baselines must be included in resource comparisons; agreement with a model optimum is not validation against physical truth.
Quantum computing uses a mathematical language that can seem unfamiliar to readers from programming, engineering and other technical backgrounds. Qubits are represented by vectors, quantum gates by matrices, measurement outcomes by eigenvalues, and probability amplitudes by complex numbers. Quantum Computing Preliminaries develops this language carefully, connecting each mathematical idea with its computational and physical meaning. The book begins with complex vector spaces and explains why quantum states require more than classical probabilities. It introduces qubits, superposition, normalisation, basis vectors and Dirac bra–ket notation, supported by worked calculations and geometric interpretation using the Bloch sphere. The discussion then turns to inner products and the geometry of quantum states. Readers learn how to calculate norms, test orthogonality, determine state overlap and apply the Born rule. Particular attention is given to complex conjugation and the distinction between the ordinary dot product and the Hermitian inner product used in quantum mechanics.The treatment of matrix operators establishes the roles of linear, unitary and Hermitian operators. It explains matrix multiplication, non-commutativity, eigenvalues, eigenvectors, reversibility and norm preservation. The Pauli and Hadamard gates provide concrete examples that link matrix algebra to rotations, the Bloch sphere, quantum-state evolution, and measurement. Outer products and projection operators are developed as tools for constructing operators, selecting state components and describing ideal projective measurements. The book examines idempotence, completeness, repeated measurement and the distinction between the ket–bra outer product and the tensor product used to form composite quantum systems. The final chapter explores changes of basis, operator representations and diagonalisation. It shows how the same physical state or operation can have different numerical representations in the computational and Hadamard bases. Readers derive the similarity transformation, use outer-product substitution, diagonalise Hermitian operators and interpret the spectral decomposition geometrically. Examples involving the Pauli-X, Pauli-Y, Pauli-Z and Hadamard operators demonstrate how basis choice exposes phase, population and measurement structure without changing the underlying physics. Throughout the book, Python and NumPy examples translate the mathematics into executable calculations. The programs demonstrate complex-vector operations, inner products, eigendecomposition, basis transformation and numerical verification while explaining the importance of using operations suited to complex Hilbert spaces. Each chapter includes conceptual, mathematical, computational and challenge exercises. Detailed solutions provide step-by-step derivations, program output and physical interpretation, allowing the book to support both guided study and independent practice. Quantum Computing Preliminaries is intended for students, software developers, engineers and technically minded readers preparing to study quantum circuits, algorithms, simulation, measurement, entanglement or error correction. It assumes familiarity with elementary algebra and programming but does not require prior training in quantum mechanics. Its purpose is to provide a clear mathematical foundation for understanding how quantum information is represented, transformed and measured.
In quantum image processing, a fundamental step is encoding classical image data into quantum states. This can be achieved using methods such as Flexible Representation of Quantum Images (FRQI), Quantum Probability Image Encoding (QPIE), and Novel Enhanced Quantum Representation (NEQR). However, on real quantum hardware, these encodings can quickly lead to circuits with many gates, large circuit depth, and high qubit usage, which is a problem for Noisy Intermediate-Scale Quantum (NISQ) devices. In this work, we investigate whether low-rank state approximation, formulated via Schmidt decomposition, can help reduce this complexity. The method keeps only the most significant parts of a quantum state's entanglement structure, making state preparation more efficient while preserving most of the image information. We compare the three encoding techniques in their original form and with low-rank approximation, evaluating metrics such as circuit depth, CNOT count, MSE, and visual quality of reconstructed images. The results reveal meaningful trade-offs between accuracy and resource efficiency, with the FRQI model achieving a 97 percent reduction in circuit depth while maintaining a near-perfect reconstruction (MSE of about 0.27). This demonstrates the potential of low-rank techniques for advancing practical quantum image processing on near-term hardware.
This chapter examines quantum linear algebra as a foundational framework for quantum machine learning. Because many core learning tasks, including regression, dimensionality reduction, kernel evaluation, recommendation systems, and representation learning, depend on linear algebraic operations, this chapter explores how these operations can be reformulated in the quantum setting. Emphasis is placed on the shift from explicit numerical computation to coherent manipulation of quantum states and measurement-based access to information. This chapter introduces quantum inner product estimation as a primitive for similarity and kernel evaluation, highlighting the swap test and related overlap-estimation methods. It then develops quantum matrix inversion and the HHL algorithm, emphasizing the preparation of solution states rather than explicit solution vectors and discussing the importance of conditioning and regularization. Quantum singular value decomposition is presented through Hermitian embeddings, phase estimation, block encoding, and quantum singular value transformation, providing a unified framework for spectral filtering and low-rank approximation. This chapter concludes with quantum principal component analysis, in which covariance structure is represented by a density matrix and accessed through quantum phase estimation and density matrix exponentiation. Throughout, practical demonstrations in Qiskit and PennyLane illustrate the algorithms, while careful attention is given to the limitations imposed by state preparation, measurement, noise, and classical data loading.
This chapter introduces quantum computing as a new computational paradigm built on superposition, entanglement, interference, and measurement and establishes its relevance to quantum searching and optimization. It first defines the core information primitives—qubits, quantum gates, and quantum circuits—and explains how they enable nonclassical state manipulation compared with deterministic bit-based processing. This chapter then surveys three major models of quantum computation: the gate-based circuit model for algorithmic execution via unitary gate sequences, the measurement-based model using pre-entangled resource states with computation driven by adaptive measurements, and the adiabatic model that encodes solutions in the ground state of a gradually evolved system, aligning naturally with annealing-style optimization. A structured comparison between classical and quantum computing highlights differences in data representation, intrinsic parallelism, scalability constraints, and error behavior and situates quantum supremacy as a task-specific benchmark rather than a universal replacement for classical machines. This chapter connects these foundations to application-critical algorithms, emphasizing Grover’s search and Shor’s factoring as canonical demonstrations of computational advantage. Finally, it motivates the book’s focus by mapping quantum search and optimization to practical domains including database querying, cryptography and post-quantum security, financial risk and fraud analytics, supply-chain logistics, scientific simulation, and emerging workflows such as quantum-assisted drug discovery and routing problems, while noting current hardware limitations and the need for robust error control.
A finite quantum memory is specified by more than its entanglement entropy. This paper determines how to shape its Schmidt weights and allocate optical photons to minimize the quantum conditional uncertainty of a prescribed heterodyne measurement. The resulting resource laws let researchers compare a uniformly weighted entangled source with a shaped source, and assess when sharing one memory across optical modes is more efficient than independently preparing mode–memory pairs. Complete preparation model and sharp leading resource law. Let Vm,D(E)V_{m,D}(E)Vm,D​(E) be the minimum of Hμ(Z∣B)=hμ(Z)−χ(Z:B)H_\mu(Z|B)=h_\mu(Z)-\chi(Z:B)Hμ​(Z∣B)=hμ​(Z)−χ(Z:B) over all normal, possibly mixed optical–memory states with total mean photon number at most EEE and total accessible memory dimension at most DDD. Here ZZZ is the outcome of an mmm-mode heterodyne measurement, BBB is the retained quantum memory, χ\chiχ is Holevo information, and dμ=π−md2mzd\mu=\pi^{-m}d^{2m}zdμ=π−md2mz. Accessible flags count toward the same memory budget. For every fixed mmm and every joint limit D,E→∞D,E\to\inftyD,E→∞, Vm,D(E)∼Λm(E/TD)TD,TD=(m!D)1/m.V_{m,D}(E)\sim\frac{\Lambda_m(E/T_D)}{T_D},\qquad T_D=(m!D)^{1/m}.Vm,D​(E)∼TD​Λm​(E/TD​)​,TD​=(m!D)1/m. The function Λm\Lambda_mΛm​ is an explicitly characterized constrained radial ground-state energy. Its Kummer and Bessel profiles give legal attaining preparations. The converse retains arbitrary coherent inputs and their photon tails, rather than optimizing only a diagonal ansatz. Why shaping changes the resource exponent. At the common photon ceiling E=TDE=T_DE=TD​, optimally shaped sources have uncertainty of order D−1/mD^{-1/m}D−1/m, whereas final flat Schmidt spectra have optimal order D−1/(2m)D^{-1/(2m)}D−1/(2m), even after optimizing the occupied optical subspace. Flat weights create a sharp support boundary; tapering the weights replaces its boundary-layer cost by a smaller gradient cost. To reach a small target uncertainty ε\varepsilonε, this changes the leading memory requirement from 2mlog⁡2(1/ε)2m\log_2(1/\varepsilon)2mlog2​(1/ε) to mlog⁡2(1/ε)m\log_2(1/\varepsilon)mlog2​(1/ε), and the attainable photon scaling from quadratic to linear in 1/ε1/\varepsilon1/ε. This improvement is compatible with nearly maximal entanglement entropy. For the explicit canonical saturated construction, S(ρB,D)=ln⁡D−cm+o(1),∥ρB,D−ID/D∥1⟶Lm>0.S(\rho_{B,D})=\ln D-c_m+o(1),\qquad \|\rho_{B,D}-I_D/D\|_1\longrightarrow L_m>0.S(ρB,D​)=lnD−cm​+o(1),∥ρB,D​−ID​/D∥1​⟶Lm​>0. The full trace-norm limits are approximately L1=0.8771L_1=0.8771L1​=0.8771 and L2=1.1170L_2=1.1170L2​=1.1170; the half-trace-distance convention halves these numbers. Thus a vanishing relative entropy deficit does not make the optimized spectrum uniform. Collective preparation at fully counted resources. For two modes in the energy-sufficient asymptotic regime, sharing quantum memory reduces the leading conditional entropy by about 10.24%10.24\%10.24% relative to the optimized independent mode–memory preparation class. That comparator includes local state optimization, integer memory and energy allocations, mixtures and charged accessible flags; it is not the class of all separable joint states. A separate finite construction uses seven coherent optical inputs, three retained qubits and 0.210.210.21 input photons. Under the specified loss, phase and stochastic memory-fault contract, its retained photon budget is at most 0.2007810.2007810.200781, and its conditional entropy is strictly lower than that of the same independent comparator by more than 0.005475350.005475350.00547535 nats. The integrated article gives the preparation circuit, one-sided numerical bounds and complete same-source error accounting. Interpretation. The objective uses quantum side information. A classical measurement of the memory generally has accessible information below χ(Z:B)\chi(Z:B)χ(Z:B), so the conditional-entropy advantages do not by themselves assert the same improvement in estimation precision or classical readout performance. The finite circuit is a rigorously specified source model, not a qualification of an implemented laboratory apparatus. The single PDF contains the article, five figures, full proofs and references.
How much internal quantum memory is needed to reproduce a stationary stochastic process, and how does reducing that memory change the environmental resources required when output records are retained? This paper establishes exact quantum-source constructions and resource bounds under explicitly specified readout and retained-output conditions. A general completion theorem identifies the minimum Hilbert-space dimension for stationary triple data in the orthogonal-current-symbol source class: it equals the sum of the complex positive-semidefinite ranks of the middle-symbol slices. An attaining construction reproduces the entire canonical stationary order-two process. For the delayed trine process, the minimum classical hidden Markov model requires nine states at every non-independent parameter value. Quantum instruments reproduce the same complete output law using four-dimensional memory throughout the parameter range, with four dimensions proved necessary in a certified low-noise regime and three dimensions both sufficient and necessary on an explicit high-noise interval. The paper then separates reusable memory dimension from environmental dilation cost. On the certified high-noise interval, the exact one-step Pareto frontier consists of a three-dimensional source with recorded-channel Choi rank nine and a four-dimensional source with rank three. Here, Choi rank measures the minimum initially pure environmental remainder dimension for the channel retaining both the emitted classical symbol and final quantum memory. A separate mixed-support qutrit construction at noise parameter 0.99 improves complete-history costs and gives upper bounds close to universal lower bounds. These results show why optimizing a single step need not optimize a retained multi-step history. The analysis also distinguishes complete channels, fixed-input states and archive-only outputs, and demonstrates how access to a retained reference can distinguish generators with identical passive output statistics. Full-word proofs, algebraic complete-positivity certificates, a technical companion and reproducibility materials accompany the results. The exact intermediate memory boundary and complete-history optimum remain open. All results use standard quantum probabilities and apply to the stated operational resource models; they do not establish total hardware, energy or thermodynamic minima.