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:2610.03393

Multivariate quantum signal processing with optimal query complexity

Mingrui Jing, Xin Wang, Zhan Yu, Lei Zhang

On Aqora:Zhan YuZhan Yu
quant-phSubmitted Oct 2, 2026• Announced Oct 5, 2026• 1 citation
arXivHTMLPDF

Abstract

Polynomial transformations of data encoded in quantum operations are a building block of quantum algorithms, and their query complexity measures how often those operations are used. For multiple variables, existing methods either realize restricted polynomial families or implement monomials separately at a query cost exponential in the number of variables. We present a quantum circuit that implements every multivariate trigonometric polynomial up to an explicit normalization. The circuit processes one variable coherently over the frequency components of the remaining variables, allowing all components to share the same signal queries. Each variable is queried exactly as many times as its degree in the polynomial, and we prove that this query complexity is optimal. We further extend the construction to commuting unitaries. The resulting circuit applies the multivariate polynomial simultaneously across their shared eigenbasis, using the minimum numbers of forward and inverse queries to each unitary. We also study trainable rotations with fixed preparation and a local readout of the last input. At query depth at least two, independent uniform initialization gives every active readout gradient a variance lower bound set by its squared branch weight and the inverse square root of the depth, uniformly over data states and dimensions. For supervised squared loss, retaining sample correlations gives an initialization gradient-variance bound in terms of targets and input spectra. This also bounds the expected loss decrease after the first exact gradient step with a specified step size. These results provide a scalable and practical solution for developing multivariate quantum algorithms and quantum learning models.

AI Summary

The paper asks whether general multivariate polynomial transformations can attain degree-matched query counts. It constructs exact block encodings of trigonometric polynomials with prescribed parity and coefficient norm $\|P\|_{c,1}\leq1$, using exactly $L_j$ plain $R_z(x_j)$ queries per variable. Frequency labels for the first $N-1$ variables coherently control univariate QSP sequences in the last variable, sharing signal queries across branches. With $L_N=\max_jL_j$ and $\ell=\prod_{j<N}(L_j+1)$, the construction uses at most $2(L_N+2)\ell-2$ polynomial-dependent rotation parameters. Exponent bookkeeping proves matching exact-query lower bounds. For commuting unitaries accessed through controlled forward and inverse oracles, the construction implements Laurent polynomials using respectively $\deg_j^+(P)$ and $\deg_j^-(P)$ queries, simultaneously attaining both lower bounds. An appendix characterizes optimal normalization at a fixed scalar-query order by an SDP, with circuit recovery and dual certificates, and exhibits a target whose optimal normalization changes from $1$ to $\sqrt2$ when query order changes. For fixed preparation and a local $Z$ readout, direct two-copy gate averaging gives initialization gradient variance at least $|a_\mu|^4/(64\sqrt{2L_N})$ for every active coordinate when $L_N\geq2$; the depth exponent is sharp. Related squared-loss bounds retain sample correlations. The normalization requirement is stronger than supremum-norm boundedness, and parameter counts remain exponential in variable number for dense polynomials. Unitary inputs must commute. The learning readout depends only on the last input; uniform branch weights produce exponentially small coordinate variances in the number of inputs. Zero-loss convergence requires an additional trajectory-wide Jacobian condition, not established by initialization bounds.

Little sign of AI-generated textUnclear noveltyStrong assumptions

Central claim may not hold: The abstract claims implementation of every multivariate trigonometric polynomial up to normalization, but the stated scalar query model cannot implement mixed-parity targets such as (1+e^{ix/2})/2 at any fixed query count, even after nonzero constant rescaling. The preliminaries incorrectly dismiss parity as a restriction on realizable functions: doubling integer-frequency exponents does not encompass the mixed-parity polynomials included in the paper's definition.

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

Comments
30 pages including appendix

Related papers

Quantum Algorithms for Multivariable Polynomial Transformations: From Efficient Synthesis to Quantum Channel Transformations

Zheyu Shen, Yusen Wu, Xiao Yuan +2 more

Polynomial transformations are basic primitives in quantum algorithms: quantum signal processing and singular value transformation compile univariate polynomials into circuits with query complexity essentially set by degree. Multivariable transformations of noncommuting matrices, however, lack a comparable synthesis theory. We develop a complete constructive theory under joint block access for all polynomials contractive on the prescribed matrix domain. Given a compact finite-state description of a degree-$D$ polynomial $P$ and $0<τ<1$, we synthesize $P$ with $(1+τ)$-optimal normalization using $O(D/\sqrtτ)$ queries, and exactly $D$ under row-block access, matching a degree lower bound. The complete circuit realization is classically computable in polynomial time with polylogarithmic dependence on accuracy. The key to our construction is a finite algorithmic Schur--Agler theorem: the coefficient recurrence defines a polynomial-dimensional continuation space supporting a complete semidefinite certificate for contractivity on the prescribed domain. Factoring this certificate yields the contractive realization underlying the query algorithm. Beyond matrix transformations, our framework lifts multivariable polynomial synthesis to quantum channel transformations. Given coherent Kraus access $K=\{K_a\}_a$, we synthesize any finite jointly contractive family of noncommutative polynomial maps $K\mapsto\{F_b(K)\}_b$ as completely positive operations, allowing coherent interference among Kraus histories. Furthermore, channel-level transformations specified by causal Choi data can be synthesized explicitly as fixed-order quantum combs. Collectively, these results point to a broader program: multivariable approximation as a language for multi-operator quantum algorithms and higher-order quantum information processing.

quant-pharXiv:2610.08714•Oct 7, 2026•1 citation

Multivariate Quantum Signal Processing

Guang Hao Low

We develop Multivariate Quantum Signal Processing for designing matrix-valued transformations of multiple noncommuting nonnormal block-encoded matrices. Our framework equips quantum transducers with the analytic structure of multiport Infinite-Impulse-Response (IIR) filters and establishes a transfer-function methodology for designing quantum query algorithms. Efficiently computable analytic bounds that hold uniformly across the oracle promise yield efficient quantum circuits with logarithmic error dependence and near-unity success probability, paralleling the approximation-theoretic guarantees of univariate QSP/QSVT. We provide a complete characterization of achievable transfer functions and a principled library of analytic IIR components. Our modular designs greatly improve or resolve the per-oracle query complexity of open multi-oracle problems: (1) Hamiltonian simulation under $H=\sum_j H_j$ with weighted quasinorm cost $\approx t\langle\vecλ,\vec C\rangle_{1/2}+\mathcal{O}(\|\vec C\|_1\log(1/ε))$ for block-encoding normalizations $\vecλ$ and costs $\vec C$; (2) sparse simulation with optimal time--norm $Θ(\sqrt{d}\|H\|_{\rm 1\to2}t)$ and additive error dependence; (3) solving linear systems and (4) differential equations under minimal matrix assumptions with optimal weighted state-preparation and multi-term cost; (5) ground-state preparation and energy estimation; (6) generalized eigenvalue problems. We apply these to exactly simulate downfolded Hamiltonians with leading cost $t\langle\vecλ_{\rm eff},\vec C_{\rm eff}\rangle_{1/2}$ further reduced by fractional dynamic correlation energy contributions from coarse-grained Hamiltonians, and show similar classical-data-input-reduction results for multiscale graph classification via Kron reduction. We also show the online generalization with anisotropic Kalman filtering of streaming quantum data.

quant-pharXiv:2610.01125•Oct 2, 2026•2 citations

Simulation of Non-Hermitian Hamiltonians with Bivariate Quantum Signal Processing

Joshua M. Courtney

We achieve query-optimal quantum simulations of non-Hermitian Hamiltonians $H_{\mathrm{eff}} = H_R + iH_I$, where $H_R$ is Hermitian and $H_I \succeq 0$, using a bivariate extension of quantum signal processing (QSP) with non-commuting signal operators. The algorithm encodes the interaction-picture Dyson series as a polynomial on the bitorus, implemented through a structured multivariable QSP (M-QSP) circuit. A constant-ratio condition guarantees scalar angle-finding for M-QSP circuits with arbitrary non-commuting signal operators. A degree-preserving sum-of-squares spectral factorization permits scalar complementary polynomials in two variables. Angles are deterministically calculated in a classical precomputation step, running in $\mathcal{O}(d_R \cdot d_I)$ classical operations. Operator norms $α_R\,,β_I$ contribute additively with query complexity $\mathcal{O}((α_R + β_I)T + \log(1/\varepsilon)/\log\log(1/\varepsilon))$ matching an information-theoretic lower bound in the separate-oracle model, where $H_R$ and $H_I$ are accessed through independent block encodings. The postselection success probability is $e^{-2β_I T}\|e^{-iH_{\mathrm{eff}}T}|ψ_0\rangle\|^2\cdot (1 - \mathcal{O}(\varepsilon))$, decomposing into a state-dependent factor $\|e^{-iH_{\mathrm{eff}}T}|ψ_0\rangle\|^2$ from the intrinsic barrier and an $e^{-2β_I T}$ overhead from polynomial block-encoding.

quant-pharXiv:2605.12450•May 13, 2026•1 citation

Global Optimization for Parametrized Quantum Circuits

Iosif Sakos, Antonios Varvitsiotis, Georgios Korpas +1 more

In the absence of error correction, noisy intermediate-scale quantum devices are operated by training parametrized quantum circuits (PQCs) so as to minimize a suitable loss function. Finding the optimal parameters of those circuits is a hard optimization problem, where global guarantees are known only for highly structured cases of limited practical relevance, and first-order methods can fail to find even local minima due to the presence of barren plateaus. In this work, we study the training of practical classes of PQCs, namely polynomial-depth circuits with a constant number of trainable parameters. This captures widely used PQC families, including fixed-depth QAOA, hardware-efficient ansätze, and Fixed Parameter Count QAOA. Our main technical result is a fully polynomial randomized approximation scheme (FPRAS), which, for every $ε>0$, returns an $ε$-approximate solution to the problem's global optimum with high probability, and has runtime and query complexity polynomial in $1/ε$ and the number of qubits. Unlike the standard hybrid quantum-classical training loop in variational algorithms, where the quantum device is queried repeatedly throughout the training, our approach separates the computation into two distinct stages: (1) an initial quantum data-acquisition phase, followed by (2) a classical global-optimization phase based on the trigonometric moment/sum-of-squares hierarchies. Under a standard flat-extension condition, which can be checked numerically, the method also supports the extraction of optimal circuit parameters. The existence of an FPRAS implies that the promise problem associated with the optimization of poly-depth constant-parameter PQC is in BQP. This imposes a limitation on the expressive power of the class, namely, it cannot encode combinatorial optimization problems whose objective values are separated by an inverse-polynomial gap.

quant-pharXiv:2603.21757•Mar 24, 2026•2 citations

An adversary bound for quantum signal processing

Lorenzo Laneve

Quantum signal processing (QSP) and quantum singular value transformation (QSVT), have emerged as unifying frameworks in the context of quantum algorithm design. These techniques allow to carry out efficient polynomial transformations of matrices block-encoded in unitaries, involving a single ancilla qubit. Recent efforts try to extend QSP to the multivariate setting (M-QSP), where multiple matrices are transformed simultaneously. However, this generalization faces problems not encountered in the univariate counterpart: in particular, the class of polynomials achievable by M-QSP seems hard to characterize. In this work we borrow tools from query complexity, namely the state conversion problem and the adversary bound: we first recast QSP as a state conversion problem over the Hilbert space of square-integrable functions. We then show that the adversary bound for a state conversion problem in this space precisely identifies all and only the QSP protocols in the univariate case. Motivated by this first result, we extend the formalism to several variables: the existence of a feasible solution to the adversary bound implies the existence of a M-QSP protocol, and the computation of a protocol of minimal space is reduced to a rank minimization problem involving the feasible solution space of the adversary bound.

arXiv:2506.20484•Quantum•Mar 13, 2026•7 citations

Optimal Bounds, Barriers, and Extensions for Non-Hermitian Bivariate Quantum Signal Processing

Joshua M. Courtney

Multivariate quantum signal processing (M-QSP) has recently been shown to be applicable for non-Hermitian Hamiltonian simulation, opening several problems regarding the optimization landscape, angle-finding, and constant-factor analysis. We resolve several of these problems here. We find the anti-Hermitian query complexity $d_I = Θ(\betaI T + \log(1/\varepsilon)/\log\log(1/\varepsilon))$ to be tight, established via Chebyshev coefficient bounds, modified Bessel function asymptotics, and Lambert~$W$ inversion. Fast-forwarding to $d_I = \mathcal{O}(\sqrt{\betaI T})$ is impossible in the bivariate polynomial model, though a linear state-dependent improvement to $d_I = \mathcal{O} β_{\mathrm{eff}} T + \log(1/\varepsilon)/\log\log(1/\varepsilon))$ is achievable. The optimization landscape of M-QSP admits spurious local minima, but a warm-start basin guarantee ensures the two-stage algorithm converges. CRC-exploiting block peeling reduces angle-finding from $\mathcal{O}(d^3)$ to $\mathcal{O}(d^2)$ classical operations, and optimized error allocation yields a leading constant of approximately~$2$ relative to the information-theoretic lower bound. A constant-ratio condition extends to non-identical signal operators, enabling time-dependent non-Hermitian simulation with query complexity $\mathcal{O}(\int_0^T(\alphaR(s) + \betaI(s))\,ds + \log(1/\varepsilon)/\log\log(1/\varepsilon))$. Block-encoding overhead $e^{-2\betaI T}$ holds across all function classes within the walk-operator oracle model, and dilational methods (Schrödingerization) achieve the walk-operator barrier. A precisely characterized direct-access construction achieves the intrinsic barrier $e^{-2ωT}$ (with $ω< \betaI$ for non-commuting Hamiltonians) on a restricted domain, though extension to the full bitorus remains open.

quant-pharXiv:2605.12656•May 14, 2026•0 citations
Polynomial transformations of data encoded in quantum operations are a building block of quantum algorithms, and their query complexity measures how often those operations are used. For multiple variables, existing methods either realize restricted polynomial families or implement monomials separately at a query cost exponential in the number of variables. We present a quantum circuit that implements every multivariate trigonometric polynomial up to an explicit normalization. The circuit processes one variable coherently over the frequency components of the remaining variables, allowing all components to share the same signal queries. Each variable is queried exactly as many times as its degree in the polynomial, and we prove that this query complexity is optimal. We further extend the construction to commuting unitaries. The resulting circuit applies the multivariate polynomial simultaneously across their shared eigenbasis, using the minimum numbers of forward and inverse queries to each unitary. We also study trainable rotations with fixed preparation and a local readout of the last input. At query depth at least two, independent uniform initialization gives every active readout gradient a variance lower bound set by its squared branch weight and the inverse square root of the depth, uniformly over data states and dimensions. For supervised squared loss, retaining sample correlations gives an initialization gradient-variance bound in terms of targets and input spectra. This also bounds the expected loss decrease after the first exact gradient step with a specified step size. These results provide a scalable and practical solution for developing multivariate quantum algorithms and quantum learning models.
The paper asks whether general multivariate polynomial transformations can attain degree-matched query counts. It constructs exact block encodings of trigonometric polynomials with prescribed parity and coefficient norm ∥P∥c,1≤1\|P\|_{c,1}\leq1∥P∥c,1​≤1, using exactly LjL_jLj​ plain Rz(xj)R_z(x_j)Rz​(xj​) queries per variable. Frequency labels for the first N−1N-1N−1 variables coherently control univariate QSP sequences in the last variable, sharing signal queries across branches. With LN=max⁡jLjL_N=\max_jL_jLN​=maxj​Lj​ and ℓ=∏j<N(Lj+1)\ell=\prod_{j<N}(L_j+1)ℓ=∏j<N​(Lj​+1), the construction uses at most 2(LN+2)ℓ−22(L_N+2)\ell-22(LN​+2)ℓ−2 polynomial-dependent rotation parameters. Exponent bookkeeping proves matching exact-query lower bounds.
For commuting unitaries accessed through controlled forward and inverse oracles, the construction implements Laurent polynomials using respectively deg⁡j+(P)\deg_j^+(P)degj+​(P) and deg⁡j−(P)\deg_j^-(P)degj−​(P) queries, simultaneously attaining both lower bounds. An appendix characterizes optimal normalization at a fixed scalar-query order by an SDP, with circuit recovery and dual certificates, and exhibits a target whose optimal normalization changes from 111 to 2\sqrt22​ when query order changes.
For fixed preparation and a local ZZZ readout, direct two-copy gate averaging gives initialization gradient variance at least ∣aμ∣4/(642LN)|a_\mu|^4/(64\sqrt{2L_N})∣aμ​∣4/(642LN​​) for every active coordinate when LN≥2L_N\geq2LN​≥2; the depth exponent is sharp. Related squared-loss bounds retain sample correlations.
The normalization requirement is stronger than supremum-norm boundedness, and parameter counts remain exponential in variable number for dense polynomials. Unitary inputs must commute. The learning readout depends only on the last input; uniform branch weights produce exponentially small coordinate variances in the number of inputs. Zero-loss convergence requires an additional trajectory-wide Jacobian condition, not established by initialization bounds.
Polynomial transformations are basic primitives in quantum algorithms: quantum signal processing and singular value transformation compile univariate polynomials into circuits with query complexity essentially set by degree. Multivariable transformations of noncommuting matrices, however, lack a comparable synthesis theory. We develop a complete constructive theory under joint block access for all polynomials contractive on the prescribed matrix domain. Given a compact finite-state description of a degree-DDD polynomial PPP and 0<τ<10<τ<10<τ<1, we synthesize PPP with (1+τ)(1+τ)(1+τ)-optimal normalization using O(D/τ)O(D/\sqrtτ)O(D/τ​) queries, and exactly DDD under row-block access, matching a degree lower bound. The complete circuit realization is classically computable in polynomial time with polylogarithmic dependence on accuracy. The key to our construction is a finite algorithmic Schur--Agler theorem: the coefficient recurrence defines a polynomial-dimensional continuation space supporting a complete semidefinite certificate for contractivity on the prescribed domain. Factoring this certificate yields the contractive realization underlying the query algorithm. Beyond matrix transformations, our framework lifts multivariable polynomial synthesis to quantum channel transformations. Given coherent Kraus access K={Ka}aK=\{K_a\}_aK={Ka​}a​, we synthesize any finite jointly contractive family of noncommutative polynomial maps K↦{Fb(K)}bK\mapsto\{F_b(K)\}_bK↦{Fb​(K)}b​ as completely positive operations, allowing coherent interference among Kraus histories. Furthermore, channel-level transformations specified by causal Choi data can be synthesized explicitly as fixed-order quantum combs. Collectively, these results point to a broader program: multivariable approximation as a language for multi-operator quantum algorithms and higher-order quantum information processing.
We develop Multivariate Quantum Signal Processing for designing matrix-valued transformations of multiple noncommuting nonnormal block-encoded matrices. Our framework equips quantum transducers with the analytic structure of multiport Infinite-Impulse-Response (IIR) filters and establishes a transfer-function methodology for designing quantum query algorithms. Efficiently computable analytic bounds that hold uniformly across the oracle promise yield efficient quantum circuits with logarithmic error dependence and near-unity success probability, paralleling the approximation-theoretic guarantees of univariate QSP/QSVT. We provide a complete characterization of achievable transfer functions and a principled library of analytic IIR components. Our modular designs greatly improve or resolve the per-oracle query complexity of open multi-oracle problems: (1) Hamiltonian simulation under H=∑jHjH=\sum_j H_jH=∑j​Hj​ with weighted quasinorm cost ≈t⟨λ⃗,C⃗⟩1/2+O(∥C⃗∥1log⁡(1/ε))\approx t\langle\vecλ,\vec C\rangle_{1/2}+\mathcal{O}(\|\vec C\|_1\log(1/ε))≈t⟨λ,C⟩1/2​+O(∥C∥1​log(1/ε)) for block-encoding normalizations λ⃗\vecλλ and costs C⃗\vec CC; (2) sparse simulation with optimal time--norm Θ(d∥H∥1→2t)Θ(\sqrt{d}\|H\|_{\rm 1\to2}t)Θ(d​∥H∥1→2​t) and additive error dependence; (3) solving linear systems and (4) differential equations under minimal matrix assumptions with optimal weighted state-preparation and multi-term cost; (5) ground-state preparation and energy estimation; (6) generalized eigenvalue problems. We apply these to exactly simulate downfolded Hamiltonians with leading cost t⟨λ⃗eff,C⃗eff⟩1/2t\langle\vecλ_{\rm eff},\vec C_{\rm eff}\rangle_{1/2}t⟨λeff​,Ceff​⟩1/2​ further reduced by fractional dynamic correlation energy contributions from coarse-grained Hamiltonians, and show similar classical-data-input-reduction results for multiscale graph classification via Kron reduction. We also show the online generalization with anisotropic Kalman filtering of streaming quantum data.
We achieve query-optimal quantum simulations of non-Hermitian Hamiltonians Heff=HR+iHIH_{\mathrm{eff}} = H_R + iH_IHeff​=HR​+iHI​, where HRH_RHR​ is Hermitian and HI⪰0H_I \succeq 0HI​⪰0, using a bivariate extension of quantum signal processing (QSP) with non-commuting signal operators. The algorithm encodes the interaction-picture Dyson series as a polynomial on the bitorus, implemented through a structured multivariable QSP (M-QSP) circuit. A constant-ratio condition guarantees scalar angle-finding for M-QSP circuits with arbitrary non-commuting signal operators. A degree-preserving sum-of-squares spectral factorization permits scalar complementary polynomials in two variables. Angles are deterministically calculated in a classical precomputation step, running in O(dR⋅dI)\mathcal{O}(d_R \cdot d_I)O(dR​⋅dI​) classical operations. Operator norms αR ,βIα_R\,,β_IαR​,βI​ contribute additively with query complexity O((αR+βI)T+log⁡(1/ε)/log⁡log⁡(1/ε))\mathcal{O}((α_R + β_I)T + \log(1/\varepsilon)/\log\log(1/\varepsilon))O((αR​+βI​)T+log(1/ε)/loglog(1/ε)) matching an information-theoretic lower bound in the separate-oracle model, where HRH_RHR​ and HIH_IHI​ are accessed through independent block encodings. The postselection success probability is e−2βIT∥e−iHeffT∣ψ0⟩∥2⋅(1−O(ε))e^{-2β_I T}\|e^{-iH_{\mathrm{eff}}T}|ψ_0\rangle\|^2\cdot (1 - \mathcal{O}(\varepsilon))e−2βI​T∥e−iHeff​T∣ψ0​⟩∥2⋅(1−O(ε)), decomposing into a state-dependent factor ∥e−iHeffT∣ψ0⟩∥2\|e^{-iH_{\mathrm{eff}}T}|ψ_0\rangle\|^2∥e−iHeff​T∣ψ0​⟩∥2 from the intrinsic barrier and an e−2βITe^{-2β_I T}e−2βI​T overhead from polynomial block-encoding.
In the absence of error correction, noisy intermediate-scale quantum devices are operated by training parametrized quantum circuits (PQCs) so as to minimize a suitable loss function. Finding the optimal parameters of those circuits is a hard optimization problem, where global guarantees are known only for highly structured cases of limited practical relevance, and first-order methods can fail to find even local minima due to the presence of barren plateaus. In this work, we study the training of practical classes of PQCs, namely polynomial-depth circuits with a constant number of trainable parameters. This captures widely used PQC families, including fixed-depth QAOA, hardware-efficient ansätze, and Fixed Parameter Count QAOA. Our main technical result is a fully polynomial randomized approximation scheme (FPRAS), which, for every ε>0ε>0ε>0, returns an εεε-approximate solution to the problem's global optimum with high probability, and has runtime and query complexity polynomial in 1/ε1/ε1/ε and the number of qubits. Unlike the standard hybrid quantum-classical training loop in variational algorithms, where the quantum device is queried repeatedly throughout the training, our approach separates the computation into two distinct stages: (1) an initial quantum data-acquisition phase, followed by (2) a classical global-optimization phase based on the trigonometric moment/sum-of-squares hierarchies. Under a standard flat-extension condition, which can be checked numerically, the method also supports the extraction of optimal circuit parameters. The existence of an FPRAS implies that the promise problem associated with the optimization of poly-depth constant-parameter PQC is in BQP. This imposes a limitation on the expressive power of the class, namely, it cannot encode combinatorial optimization problems whose objective values are separated by an inverse-polynomial gap.
Quantum signal processing (QSP) and quantum singular value transformation (QSVT), have emerged as unifying frameworks in the context of quantum algorithm design. These techniques allow to carry out efficient polynomial transformations of matrices block-encoded in unitaries, involving a single ancilla qubit. Recent efforts try to extend QSP to the multivariate setting (M-QSP), where multiple matrices are transformed simultaneously. However, this generalization faces problems not encountered in the univariate counterpart: in particular, the class of polynomials achievable by M-QSP seems hard to characterize. In this work we borrow tools from query complexity, namely the state conversion problem and the adversary bound: we first recast QSP as a state conversion problem over the Hilbert space of square-integrable functions. We then show that the adversary bound for a state conversion problem in this space precisely identifies all and only the QSP protocols in the univariate case. Motivated by this first result, we extend the formalism to several variables: the existence of a feasible solution to the adversary bound implies the existence of a M-QSP protocol, and the computation of a protocol of minimal space is reduced to a rank minimization problem involving the feasible solution space of the adversary bound.
Multivariate quantum signal processing (M-QSP) has recently been shown to be applicable for non-Hermitian Hamiltonian simulation, opening several problems regarding the optimization landscape, angle-finding, and constant-factor analysis. We resolve several of these problems here. We find the anti-Hermitian query complexity $d_I = Θ(\betaI T + \log(1/\varepsilon)/\log\log(1/\varepsilon))$ to be tight, established via Chebyshev coefficient bounds, modified Bessel function asymptotics, and Lambert~WWW inversion. Fast-forwarding to $d_I = \mathcal{O}(\sqrt{\betaI T})$ is impossible in the bivariate polynomial model, though a linear state-dependent improvement to dI=OβeffT+log⁡(1/ε)/log⁡log⁡(1/ε))d_I = \mathcal{O} β_{\mathrm{eff}} T + \log(1/\varepsilon)/\log\log(1/\varepsilon))dI​=Oβeff​T+log(1/ε)/loglog(1/ε)) is achievable. The optimization landscape of M-QSP admits spurious local minima, but a warm-start basin guarantee ensures the two-stage algorithm converges. CRC-exploiting block peeling reduces angle-finding from O(d3)\mathcal{O}(d^3)O(d3) to O(d2)\mathcal{O}(d^2)O(d2) classical operations, and optimized error allocation yields a leading constant of approximately~222 relative to the information-theoretic lower bound. A constant-ratio condition extends to non-identical signal operators, enabling time-dependent non-Hermitian simulation with query complexity $\mathcal{O}(\int_0^T(\alphaR(s) + \betaI(s))\,ds + \log(1/\varepsilon)/\log\log(1/\varepsilon))$. Block-encoding overhead $e^{-2\betaI T}$ holds across all function classes within the walk-operator oracle model, and dilational methods (Schrödingerization) achieve the walk-operator barrier. A precisely characterized direct-access construction achieves the intrinsic barrier e−2ωTe^{-2ωT}e−2ωT (with $ω< \betaI$ for non-commuting Hamiltonians) on a restricted domain, though extension to the full bitorus remains open.