Pith. sign in

REVIEW 3 major objections 7 minor 1 cited by

LCQNN: Linear Combination of Quantum Neural Networks

T0 review · 3 major / 7 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This paper introduces the LCQNN, a quantum neural network built from a learnable linear combination of unitaries, and proves gradient-variance bounds that scale polynomially with local size rather than exponentially with total system…

desk verdict A genuinely new QNN architecture with a trainability proof that mishandles the shared coefficient-layer rotations; worth refereeing, but the Theorem 1 α-gradient bound is not established. read the letter →

arxiv 2507.02832 v2 pith:FPSKRM2N submitted 2025-07-03 quant-ph

Hongshun Yao , Xia Liu , Mingrui Jing , Guangxi Li , Xin Wang This is my paper
classification quant-ph MSC 68Q1281P68 PACS 03.67.Lx
keywords linearcombinationofunitariesquantumneuralnetworksbarrenplateausgradientvariancek-localgroup-equivariantmodelstrainability-expressivitytrade-offSchur-Weylduality
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper introduces the Linear Combination of Quantum Neural Networks (LCQNN), a quantum circuit that combines a learnable coefficient layer with a control-unitary layer, and argues this structure gives an explicit, tunable trade-off between expressivity and trainability. Its central theoretical result is that when the control unitaries are sampled from local unitary 2-designs and the observable is traceless, the variance of any cost gradient scales as Poly(n)/(L2^n), where L is the number of combined unitaries, instead of the exponential scaling of a fully expressive ansatz. For k-local control unitaries the bound becomes Poly(k)/(L2^k), showing that the locality of the building blocks, not the total system size, controls gradient decay. The authors verify the scaling numerically and demonstrate the model on a four-class MNIST task, and they extend the analysis to group-action settings where trainability is governed by the largest irreducible-subspace dimension. If correct, the framework offers a mechanism for focusing quantum resources into architectures that stay trainable while retaining nontrivial expressivity.

What carries the argument

The central object is the LCQNN circuit W = ∏_{j=0}^{L-1} (C-U_j)(V⊗I), which creates a superposition over L branches, each carrying a product-form probability p_j(α) from the coefficient layer and a unitary U_j(θ_j) acting on the target register. The argument's workhorse is the tree-structured probability p_j(α): its second moment over the coefficient angles is O(1/L), while the control-unitary layer contributes the standard barren-plateau variance Poly(n)/2^n. The theorems multiply these two contributions, giving Var[∂C] ∈ O(Poly(n)/(L2^n)); the k-local and group-action versions replace the control-unitary factor by Poly(k)/2^k or Poly(log dmax)/dmax.

What would settle it

Numerically compute Var[∂C/∂α_{jk}] for a small LCQNN (for example m=2, n=3, L=2) with a traceless observable and local unitary 2-designs, using exact or sampled gradients over many initializations, and compare the scaling with Poly(n)/(L2^n); if the empirical variance shows a different L-dependence or if the square of the full derivative differs from the product E[$p_j^{2}$]E[∂^2⟨U_j O U_j⟩], the factorization in Eq. (8) fails.

Watch

Extended reading notes

Core claim

The paper claims that a quantum neural network built as a linear combination of unitaries has a gradient variance that decays polynomially in the local subsystem size rather than exponentially in the total register size. Concretely, Theorem 1 proves that for an LCQNN on m+n qubits with L combined unitaries sampled from local unitary 2-designs and a traceless observable on the n-qubit register, Var[∂C] is O(Poly(n)/(L2^n)); with k-local control unitaries this becomes O(Poly(k)/(L2^k)). The same decomposition, applied to group-invariant architectures, gives a bound governed by the maximum dimension of the relevant irreducible representation spaces. The paper treats L and the locality k as tunable design levers: increasing L sharpens expressivity while only linearly increasing gradient decay, and keeping the individual unitaries k-local prevents gradients from collapsing as n grows.

Load-bearing premise

The main proof assumes that the gradient's variance splits cleanly into a piece from the mixing coefficients and a piece from the quantum circuits being mixed, as if changing one coefficient angle affected only one branch, whereas each angle actually controls several branches of the superposition at once.

Editorial extensions

If this is right

  • Increasing the number of combined unitaries L reduces gradient variance as 1/L, so LCQNN can be made more expressive without immediately losing trainability.
  • With k-local control unitaries, gradient variance depends on k rather than on the total qubit count n, so the architecture can be scaled to larger registers while keeping gradients alive.
  • The bound gives a quantitative design objective for quantum architecture search: choose L and unitary locality to hit a target gradient variance.
  • Restricting to group-invariant subspaces of nonexponential dimension avoids barren plateaus while preserving equivariance, extending the framework to geometric QML tasks.
  • Numerical experiments on the MNIST 4-class task show accuracy improving with both L and circuit depth, supporting the practical benefit of combining multiple feature extractors.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Inference: If the variance bound survives a direct check of coefficient-angle derivatives, the same tree-structured probability decomposition could be used to bound Hessian spectra of these ansätze, giving a trainability criterion for second-order optimizers.
  • Inference: The LCQNN construction suggests an architectural design rule of thumb: fix L proportional to the target accuracy and keep each block k-local, so training resources need not grow with register size n; this is testable by scaling experiments beyond the paper's 8-qubit tests.
  • Inference: The group-action version implies a concrete design principle for equivariant QNNs: project the observable onto irreducible subspaces and exclude labels whose dimension–multiplicity product is exponential; testing this on a non-SU(2) group would probe how generally Theorem 3 holds.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 7 minor

Summary. The paper introduces LCQNN, a parameterized quantum circuit in which a coefficient layer V(α) prepares a superposition over L control branches and a control-unitary layer applies one of L unitaries U_j(θ_j) to an n-qubit register, so that the cost is a linear combination of expectation values. Theorem 1 claims that the circuit can generate any pure state of the form ⊕_j √p_j |ψ_j⟩ and that the cost gradient variance is O(Poly(n)/(L2^n)); Proposition 2 gives the k-local variant O(Poly(k)/(L2^k)), and Theorem 3 gives a group-symmetric variant with variance O(Poly(log dmax)/(L dmax)). Numerical experiments on gradient scaling and on MNIST classification are reported in support of the theoretical claims.

Significance. Conditional on the proofs being repaired, the LCQNN construction is an appealing way to interpolate between a single QNN and a fully expressive LCU, and the claimed scalings would give a practical design rule: keep k small or L large to avoid barren plateaus. The θ-parameter part of the analysis is clean, since it uses only the independence of the coefficient amplitudes from the 2-design unitaries and the standard k-local barren-plateau bound, with no fitted constants; the numerical plots in Section III do verify the k-locality and L-dependence for θ gradients. The expressivity argument is essentially a statement about the ability of V to prepare a superposition and of the U_j to prepare arbitrary target states, which is plausible. The paper is clearly written and the MNIST study is a useful sanity check. However, the unresolved α-derivative issue and the incomplete Appendix B proofs affect the main theorems, so the significance is currently prospective rather than established.

major comments (3)
  1. [II C, proof of Theorem 1, Eqs. (6)-(8)] The α-derivative variance is not computed correctly. The text asserts after Eq. (6) that ∂_{α_j} C = (∂_{α_j} p_j) f_j, which would be true only if parameter α_j affected a single branch. In the implemented tree coefficient layer, an angle at level k is shared by every leaf in its subtree, so ∂_{α_k} C = Σ_l (∂_{α_k} p_l) f_l. The branch cross terms vanish because the f_l are independent with zero mean, but the variance is then Σ_l (∂_{α_k} p_l)^2 Var[f_l] over all affected leaves, not the product E[p_j^2] E[(∂f_j)^2] used in Eq. (8). Direct integration for the Ry-tree parameterization of Eq. (4) gives Var[∂_{α} C] ≈ (3/4)^{m-1} Var[f] for a root-level angle instead of the 2^{-m} Var[f] implied by Eqs. (7)-(8); the ratio grows exponentially with m. Hence the O(Poly(n)/(L2^n)) bound for the coefficient-layer derivatives in Theorem 1 is not established. Since Section III A selects a θ_j parameter for the numerical check, the experiments do not test this part of the theorem.
  2. [Appendix B, Proposition 2 proof] The proof reduces to 'one can derive the BP theorem with respect to the k systems' and 'Similar to the proof in Theorem 1', with no actual derivation, and Eq. (S2) uses undefined notation (sys_{n_i} ∈ syst, q_i). This is load-bearing because the coefficient-layer part of Proposition 2 inherits the same shared-parameter problem identified in Theorem 1. Thus Eq. (9), O(Poly(k)/(L2^k)), is established, at best, only for the θ_j derivatives, and Proposition 2's full variance bound is not supported.
  3. [Appendix B, Theorem 3 proof] The proof is internally inconsistent. The statement defines L = |H|, but the proof replaces L by 2^t and introduces s = ⌈log dmax⌉, so the coefficient layer has 2^s branches while the sum in Eq. (S5) runs over H; the relation among s, t, L, and dmax is never fixed. The sentence 'there exists a circuit that allows the system to revert to the Hilbert space H by adjusting the trivial space' is an unproved assertion on which the connection between the direct-sum state Eq. (S5) and the original cost rests. Finally, the scaling Eθ[∂²_{θμ} ⟨0|U_μ† O U_μ|0⟩] ∈ O(Poly(d_μ m_μ)/2^{d_μ m_μ}) is stated without specifying the distribution of U_μ or deriving the dependence on the multiplicities. Therefore Eq. (14) is not established as written.
minor comments (7)
  1. [II B, Eq. (2)] L is declared in R+ but is a number of unitaries and must be an integer; the product over C-U_j should be defined precisely, including which qubits control each U_j and in what order the controlled operations are applied.
  2. [Remark 1] There is a typo 'parameted circuit'; also, the sentence 'If L = 2^m' should clarify that L is an independent hyperparameter rather than automatically determined by the number of control qubits.
  3. [II C, proof of Theorem 1] The proof switches between ∂²C and Var[∂C]; since the first moment is shown to vanish, the statement should explicitly identify Var[∂C] with E[(∂C)^2].
  4. [III A] Figs. 5 and 6 are cited but the text does not state the number of samples per point or the error-bar convention; report standard errors or confidence intervals for the variance estimates.
  5. [Corollary 4] There are typos 'Poloy' in two places (Eq. (18) and the surrounding text); the intended expression is Poly(N) or Poly(log N).
  6. [Appendix A and B] There are several typos, including 'resepctively', 'irredcible', and 'sub-sapce'; more substantively, Theorem 3 uses H both for the Hilbert space and for the subset of irrep labels, which should be renamed to avoid ambiguity.
  7. [Abstract] The abstract claims LCQNN avoids 'excessive classical simulability', but no simulability statement is proven in the main text; either add a formal statement or soften the claim.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the LCQNN variance bounds follow from independent 2-design barren-plateau results plus direct integration, and the self-citations are not load-bearing.

full rationale

The paper's central claim is Theorem 1's variance bound Var[∂C] ∈ O(Poly(n)/(L2^n)). The derivation chain is: (i) the cost function is written as C(α,θ) = Σ_j p_j(α)⟨0|U_j† O U_j|0⟩ in Eq. (6); (ii) E[p_j²] ∈ O(1/2^t) is obtained by direct integration over α in Eq. (7); (iii) E_θ[∂²_θj⟨0|U_j† O U_j|0⟩] ∈ O(Poly(n)/2^n) is imported from the independently established unitary-2-design barren-plateau results in Refs. [29,30], not from the authors' own work; and (iv) independence of α and θ yields the factorization in Eq. (8). No parameter is fitted to data and then renamed a prediction; the 1/L scaling is a computed moment of the explicitly defined coefficient layer. The expressivity part of Theorem 1 is a constructive statement—V is built from Ry gates to produce the coefficients p_j and each U_j is assumed universal—so it is definitional but it does not assume the variance conclusion. The self-citations, including Ref. [44] where LCQNN is said to degenerate to the authors' QSF framework and Refs. [19,36,38,46,47] used for context or as prior QML/BP examples, are not load-bearing for Theorem 1, Proposition 2, or Theorem 3, which rest on Refs. [29,30] and direct calculation. The proof's treatment of ∂_α C in Eq. (8) may raise a mathematical correctness question about cross-branch terms from shared rotation angles, but that is not circularity because the target variance bound is not assumed as an input. Accordingly, no circular step is exhibited.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The theoretical results rest on standard BP scaling assumptions and the 2-design assumption for the control unitaries; no free parameters are fitted to obtain the variance bounds. The group-theoretic results rely on Schur-Weyl duality and an assumed block-diagonal observable decomposition. No new physical entities are postulated.

assumptions (5)
  • domain assumption The control unitaries U(θ_j) are sampled from (local) unitary 2-designs.
    Invoked in Section II.C before Theorem 1 and in Appendix B for Proposition 2; the standard BP scaling for the control layer follows only under this ensemble assumption.
  • domain assumption The standard barren plateau scaling result E[∂²⟨0|U†OU|0⟩] ∈ O(Poly(n)/2^n) for 2-design circuits.
    Relied on in Eqs. (7) and (8) and borrowed from Refs. [19, 29, 30] as an external benchmark.
  • standard math Schur-Weyl duality and the isotypical decomposition of the Hilbert space under G = SU(2) × S_N.
    Used in Section II.E, Eqs. (10) to (16), as background representation theory.
  • domain assumption The observable O decomposes block-diagonally as ⊕ O_μ after the Schur transform (Eq. 17).
    This is an assumption restricting measurements to the selected irrep subspaces; the paper notes it is analogous to purity conditions in BP literature.
  • ad hoc to paper There exists a quantum circuit that reverts the system to the original Hilbert space H after embedding the target state in the direct-sum space.
    Statement in Appendix B, Theorem 3 proof, provided without construction or reference; it is used to argue the expressivity of the group-action LCQNN.

how reviews work

0 comments
Cite this review

Pith. "Pith review of LCQNN: Linear Combination of Quantum Neural Networks." pith.science (2026). https://pith.science/paper/FPSKRM2N

@misc{pith2026250702832,
  author       = {Pith},
  title        = {Pith review of: LCQNN: Linear Combination of Quantum Neural Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FPSKRM2N}},
  note         = {Machine review of arXiv:2507.02832}
}
abstract

Quantum neural networks combine quantum computing with advanced data-driven methods, offering promising applications in quantum machine learning. However, the optimal paradigm for balancing trainability and expressivity in QNNs remains an open question. To address this, we introduce the Linear Combination of Quantum Neural Networks (LCQNN) framework, which uses the linear combination of unitaries concept to create a tunable design that mitigates vanishing gradients without incurring excessive classical simulability. We show how specific structural choices, such as adopting $k$-local control unitaries or restricting the model to certain group-theoretic subspaces, prevent gradients from collapsing while maintaining sufficient expressivity for complex tasks. We further employ the LCQNN model to handle supervised learning tasks, demonstrating its effectiveness on real datasets. In group action scenarios, we show that by exploiting symmetry and excluding exponentially large irreducible subspaces, the model circumvents barren plateaus. Overall, LCQNN provides a novel framework for focusing quantum resources into architectures that are practically trainable yet expressive enough to tackle challenging machine learning applications.

Figures

Figures reproduced from arXiv: 2507.02832 by the authors.

Figure 1
Figure 1. Illustration of LCQNN. V (α) is the coefficient layer, which can be constructed by the Ry rotation gate. The control-unitary layer consists of at most 2 m unitary for m control qubits. function, given that the U(θj ) in LCQNN is sampled from unitary 2-design ensembles, and V (α) is constructed by a Ry rotation gates. Trainability is one of the most critical concerns in QNNs. When a QNN forms a unitary 2-design, the … view at source ↗
Figure 2
Figure 2. LCQNN with k-local unitary. The U(θjt) in the control￾unitary layer acts on k-local systems and measurement operators act on ni systems, respectively. The coefficient layer V (α) can be con￾structed by a control Ry gate. through carefully designed circuit architectures. Notably, this quantified relationship may facilitate applications in quantum architecture search (QAS) [45–47]. D. LCQNN with k-local Unitary As ill… view at source ↗
Figure 3
Figure 3. Comparison between LCQNN and traditional QNNs. a) and b) denote the conventional QNN and the LCQNN, respectively. c) [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: The k-local unitary circuit, which includes single-qubit uni￾versal gate, U3 and CNOT gates. Parameter D is the depth of this circuit [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: The trainability comparison between conventional QNNs and [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Stacking the Deck: Tunable Trainability in Stacked LCUs

    quant-ph 2026-07 conditional novelty 6.0 of 10

    Stacked LCUs of fermionic Gaussian unitaries give variance Ω(1/(n k^{3l})) against classical simulation O(k^{2l} n^3) and quantum gate count O(l k n^2), with layers l as the single dial.

Reference graph

Works this paper leans on

54 extracted references · 38 canonical work pages · cited by 1 Pith paper

  1. [1]

    We prove that its cost function’s gradient variance scales fa- vorably, inversely with the number of combined QNNs (L)

    We provide a rigorous theoretical analysis of train- ability for the LCQNN framework. We prove that its cost function’s gradient variance scales fa- vorably, inversely with the number of combined QNNs (L)

    work page

  2. [2]

    This proposition implies that LCQNN’s structure can be tailored to specific problems to guarantee trainability while retaining non-trivial quantum features

    We demonstrate the versatility of LCQNN by re- stricting it to thek-local scenario, showing that the scale of the local QNN, determines the trainabil- ity of the LCQNN framework. This proposition implies that LCQNN’s structure can be tailored to specific problems to guarantee trainability while retaining non-trivial quantum features

    work page

  3. [3]

    We generate the LCQNN framework for the group action setting, offering novel perspectives for the design of circuit structures within the geometric quantum machine learning paradigm. • Practical Validation: We demonstrate the correctness of the theoretical framework through gradient analysis experiments and the model’s potential for practical ap- plicatio...

    work page

  4. [4]

    An introduction to quantum machine learning

    Maria Schuld, Ilya Sinayskiy, and Francesco Petruccione. An introduction to quantum machine learning. Contemporary Physics, 56(2):172–185, 2015

    2015

  5. [5]

    Quantum machine learn- ing

    Jacob Biamonte, Peter Wittek, Nicola Pancotti, Patrick Reben- trost, Nathan Wiebe, and Seth Lloyd. Quantum machine learn- ing. Nature, 549(7671):195–202, 2017

    work page 2017

  6. [6]

    Toward trainability of quantum neural networks.arXiv preprint arXiv:2011.06258, 2020

    Kaining Zhang, Min-Hsiu Hsieh, Liu Liu, and Dacheng Tao. Toward trainability of quantum neural networks.arXiv preprint arXiv:2011.06258, 2020

    arXiv 2011

  7. [7]

    Recent advances for quan- tum classifiers

    Weikang Li and Dong-Ling Deng. Recent advances for quan- tum classifiers. Science China Physics, Mechanics & Astron- omy, 65(2):220301, 2022

    work page 2022

  8. [8]

    The dilemma of quantum neural networks

    Yang Qian, Xinbiao Wang, Yuxuan Du, Xingyao Wu, and Dacheng Tao. The dilemma of quantum neural networks. IEEE Transactions on Neural Networks and Learning Systems, 2022

    work page 2022

Show all 54 references
  1. [9]

    Generalization in quantum machine learning from few training data

    Matthias C Caro, Hsin-Yuan Huang, Marco Cerezo, Kunal Sharma, Andrew Sornborger, Lukasz Cincio, and Patrick J Coles. Generalization in quantum machine learning from few training data. Nature communications, 13(1):4919, 2022

    2022

  2. [10]

    Challenges and opportunities in quantum machine learning

    Marco Cerezo, Guillaume Verdon, Hsin-Yuan Huang, Lukasz Cincio, and Patrick J Coles. Challenges and opportunities in quantum machine learning. Nature Computational Science , 2(9):567–576, 2022

    2022

  3. [11]

    Recent advances for quantum neural networks in generative learning

    Jinkai Tian, Xiaoyu Sun, Yuxuan Du, Shanshan Zhao, Qing Liu, Kaining Zhang, Wei Yi, Wanrong Huang, Chaoyue Wang, Xingyao Wu, et al. Recent advances for quantum neural networks in generative learning. IEEE Transactions on Pat- tern Analysis and Machine Intelligence , 45(10):123...

    2023

  4. [12]

    Quantum machine learning beyond kernel methods

    Sofiene Jerbi, Lukas J Fiderer, Hendrik Poulsen Nautrup, Jonas M Kübler, Hans J Briegel, and Vedran Dunjko. Quantum machine learning beyond kernel methods. Nature Communica- tions, 14(1):1–8, 2023

    2023

  5. [13]

    Quantum machine learning: A hands-on tuto- rial for machine learning practitioners and researchers

    Yuxuan Du, Xinbiao Wang, Naixu Guo, Zhan Yu, Yang Qian, Kaining Zhang, Min-Hsiu Hsieh, Patrick Rebentrost, and Dacheng Tao. Quantum machine learning: A hands-on tuto- rial for machine learning practitioners and researchers. arXiv preprint arXiv:2502.01146, 2025

    2025 arXiv

  6. [14]

    Parameterized quantum circuits as machine learn- ing models

    Marcello Benedetti, Erika Lloyd, Stefan Sack, and Mattia Fiorentini. Parameterized quantum circuits as machine learn- ing models. Quantum Science and Technology , 4(4):043001, 2019

    2019

  7. [15]

    Structure optimization for parameterized quantum circuits

    Mateusz Ostaszewski, Edward Grant, and Marcello Benedetti. Structure optimization for parameterized quantum circuits. Quantum, 5:391, 2021

    2021

  8. [16]

    Quantum algorithms for quantum chemistry and quan- tum materials science

    Bela Bauer, Sergey Bravyi, Mario Motta, and Garnet Kin-Lic Chan. Quantum algorithms for quantum chemistry and quan- tum materials science. Chemical Reviews , 120(22):12685– 12717, 2020

    2020

  9. [17]

    Quantum chemistry calculations on a trapped-ion quantum simulator

    Cornelius Hempel, Christine Maier, Jonathan Romero, Jarrod McClean, Thomas Monz, Heng Shen, Petar Jurcevic, Ben P Lanyon, Peter Love, Ryan Babbush, et al. Quantum chemistry calculations on a trapped-ion quantum simulator. Physical Re- view X, 8(3):031022, 2018

    2018

  10. [18]

    A quantum approximate optimization algorithm

    Edward Farhi, Jeffrey Goldstone, and Sam Gutmann. A quantum approximate optimization algorithm. arXiv preprint arXiv:1411.4028, 2014

    2014 arXiv

  11. [19]

    Quantum approximate optimization algo- rithm: Performance, mechanism, and implementation on near- term devices

    Leo Zhou, Sheng-Tao Wang, Soonwon Choi, Hannes Pichler, and Mikhail D Lukin. Quantum approximate optimization algo- rithm: Performance, mechanism, and implementation on near- term devices. Physical Review X, 10(2):021067, 2020

    2020

  12. [20]

    Variational quan- tum circuits for deep reinforcement learning

    Samuel Yen-Chi Chen, Chao-Han Huck Yang, Jun Qi, Pin- Yu Chen, Xiaoli Ma, and Hsi-Sheng Goan. Variational quan- tum circuits for deep reinforcement learning. IEEE access , 8:141007–141024, 2020

    2020

  13. [21]

    Learning temporal data with a varia- tional quantum recurrent neural network

    Yuto Takaki, Kosuke Mitarai, Makoto Negoro, Keisuke Fujii, and Masahiro Kitagawa. Learning temporal data with a varia- tional quantum recurrent neural network. Physical Review A , 103(5):052414, 2021

    2021

  14. [22]

    Statistical analysis of quantum state learning process in quan- tum neural networks

    Hao-kai Zhang, Chenghong Zhu, Mingrui Jing, and Xin Wang. Statistical analysis of quantum state learning process in quan- tum neural networks. Advances in Neural Information Process- ing Systems, 36, 2024

    2024

  15. [23]

    Barren plateaus in quantum neural network training landscapes

    Jarrod R McClean, Sergio Boixo, Vadim N Smelyanskiy, Ryan Babbush, and Hartmut Neven. Barren plateaus in quantum neural network training landscapes. Nature communications, 9(1):4812, 2018

    2018

  16. [24]

    Cost function dependent barren plateaus in shallow parametrized quantum circuits

    Marco Cerezo, Akira Sone, Tyler V olkoff, Lukasz Cincio, and Patrick J Coles. Cost function dependent barren plateaus in shallow parametrized quantum circuits. Nature communica- tions, 12(1):1791, 2021

    2021

  17. [25]

    Exponentially many local minima in quantum neural networks

    Xuchen You and Xiaodi Wu. Exponentially many local minima in quantum neural networks. In International Conference on Machine Learning, pages 12144–12155. PMLR, 2021

    2021

  18. [26]

    Mitigating barren plateaus of variational quantum eigen- solvers

    Xia Liu, Geng Liu, Hao-Kai Zhang, Jiaxin Huang, and Xin Wang. Mitigating barren plateaus of variational quantum eigen- solvers. IEEE Transactions on Quantum Engineering, 2024

    2024

  19. [27]

    Hamiltonian variational ansatz without barren plateaus

    Chae-Yeun Park and Nathan Killoran. Hamiltonian variational ansatz without barren plateaus. Quantum, 8:1239, 2024

    2024

  20. [28]

    Explor- ing entanglement and optimization within the hamiltonian vari- ational ansatz

    Roeland Wiersema, Cunlu Zhou, Yvette de Sereville, Juan Fe- lipe Carrasquilla, Yong Baek Kim, and Henry Yuen. Explor- ing entanglement and optimization within the hamiltonian vari- ational ansatz. PRX quantum, 1(2):020319, 2020

    2020

  21. [29]

    Escaping from the barren plateau via gaussian initializations in deep variational quantum circuits

    Kaining Zhang, Liu Liu, Min-Hsiu Hsieh, and Dacheng Tao. Escaping from the barren plateau via gaussian initializations in deep variational quantum circuits. Advances in Neural Infor- mation Processing Systems, 35:18612–18627, 2022

    2022

  22. [30]

    Trainability enhancement of parameterized quantum circuits via reduced- domain parameter initialization

    Yabo Wang, Bo Qi, Chris Ferrie, and Daoyi Dong. Trainability enhancement of parameterized quantum circuits via reduced- domain parameter initialization. Physical Review Applied , 22(5):054005, 2024

    2024

  23. [31]

    Avoiding barren plateaus via gaussian mixture model

    Xiao Shi and Yun Shang. Avoiding barren plateaus via gaussian mixture model. arXiv preprint arXiv:2402.13501, 2024

    2024 arXiv

  24. [32]

    Characterizing barren plateaus in quantum ansätze with the adjoint representation

    Enrico Fontana, Dylan Herman, Shouvanik Chakrabarti, Ni- raj Kumar, Romina Yalovetzky, Jamie Heredge, Shree Hari Sureshbabu, and Marco Pistoia. Characterizing barren plateaus in quantum ansätze with the adjoint representation. Nature Communications, 15(1):7171, August 2024. Pu...

    2024

  25. [33]

    Bakalov, Frédéric Sauvage, Alexan- der F

    Michael Ragone, Bojko N. Bakalov, Frédéric Sauvage, Alexan- der F. Kemper, Carlos Ortiz Marrero, Martín Larocca, and M. Cerezo. A Lie algebraic theory of barren plateaus for deep parameterized quantum circuits. Nature Communications, 15(1):7172, August 2024. Publisher: Nature ...

    2024

  26. [34]

    Theory for equivariant quantum neural net- works

    Quynh T Nguyen, Louis Schatzki, Paolo Braccia, Michael Ragone, Patrick J Coles, Frederic Sauvage, Martin Larocca, and Marco Cerezo. Theory for equivariant quantum neural net- works. PRX Quantum, 5(2):020328, 2024

    2024

  27. [35]

    Theoretical guarantees for permutation-equivariant quantum neural networks

    Louis Schatzki, Martin Larocca, Quynh T Nguyen, Fred- eric Sauvage, and Marco Cerezo. Theoretical guarantees for permutation-equivariant quantum neural networks. npj Quan- tum Information, 10(1):12, 2024. 9

    2024

  28. [36]

    On the dynamical lie algebras of quantum approximate optimization algorithms

    Jonathan Allcock, Miklos Santha, Pei Yuan, and Shengyu Zhang. On the dynamical lie algebras of quantum approximate optimization algorithms. arXiv preprint arXiv:2407.12587 , 2024

    2024 arXiv

  29. [37]

    Analyzing the quantum approx- imate optimization algorithm: ans \" atze, symmetries, and lie algebras

    Sujay Kazi, Martín Larocca, Marco Farinati, Patrick J Coles, M Cerezo, and Robert Zeier. Analyzing the quantum approx- imate optimization algorithm: ans \" atze, symmetries, and lie algebras. arXiv preprint arXiv:2410.05187, 2024

  30. [38]

    Towards determin- ing the presence of barren plateaus in some chemically inspired variational quantum algorithms

    Rui Mao, Guojing Tian, and Xiaoming Sun. Towards determin- ing the presence of barren plateaus in some chemically inspired variational quantum algorithms. Communications Physics , 7(1):342, 2024

    2024

  31. [39]

    Quantum recurrent embedding neural network

    Mingrui Jing, Erdong Huang, Xiao Shi, Shengyu Zhang, and Xin Wang. Quantum recurrent embedding neural network. arXiv preprint arXiv:2506.13185, 2025

    2025 arXiv

  32. [40]

    Does prov- able absence of barren plateaus imply classical simulability? or, why we need to rethink variational quantum computing

    Marco Cerezo, Martin Larocca, Diego García-Martín, Nelson L Diaz, Paolo Braccia, Enrico Fontana, Manuel S Rudolph, Pablo Bermejo, Aroosa Ijaz, Supanut Thanasilp, et al. Does prov- able absence of barren plateaus imply classical simulability? or, why we need to rethink variatio...

    2023 arXiv

  33. [41]

    Power and limitations of single-qubit native quantum neural networks.Ad- vances in Neural Information Processing Systems , 35:27810– 27823, 2022

    Zhan Yu, Hongshun Yao, Mujin Li, and Xin Wang. Power and limitations of single-qubit native quantum neural networks.Ad- vances in Neural Information Processing Systems , 35:27810– 27823, 2022

    2022

  34. [42]

    Quan- tum algorithm for linear systems of equations

    Aram W Harrow, Avinatan Hassidim, and Seth Lloyd. Quan- tum algorithm for linear systems of equations. Physical review letters, 103(15):150502, 2009

    2009

  35. [43]

    Hamiltonian simulation using linear combinations of unitary operations

    Andrew M Childs and Nathan Wiebe. Hamiltonian simulation using linear combinations of unitary operations. arXiv preprint arXiv:1202.5822, 2012

    2012 arXiv

  36. [44]

    Multivariable quantum sig- nal processing (m-qsp): prophecies of the two-headed oracle

    Zane M Rossi and Isaac L Chuang. Multivariable quantum sig- nal processing (m-qsp): prophecies of the two-headed oracle. Quantum, 6:811, 2022

    2022

  37. [45]

    Guang Hao Low and Isaac L. Chuang. Optimal hamiltonian simulation by quantum signal processing. Phys. Rev. Lett. , 118:010501, Jan 2017

    2017

  38. [46]

    Quantum singular value transformation & its algorithmic applications

    András Gilyén. Quantum singular value transformation & its algorithmic applications . PhD thesis, University of Amster- dam, 2019

    2019

  39. [47]

    Nonlinear functions of quantum states

    Hongshun Yao, Yingjian Liu, Tengxiang Lin, and Xin Wang. Nonlinear functions of quantum states. arXiv preprint arXiv:2412.01696, 2024

    2024 arXiv

  40. [48]

    Quan- tum architecture search: a survey

    Darya Martyniuk, Johannes Jung, and Adrian Paschke. Quan- tum architecture search: a survey. In 2024 IEEE International Conference on Quantum Computing and Engineering (QCE) , volume 1, pages 1695–1706. IEEE, 2024

    2024

  41. [49]

    Predicting quantum learnability from landscape fluctuation.arXiv preprint arXiv:2406.11805, 2024

    Hao-Kai Zhang, Chenghong Zhu, and Xin Wang. Predicting quantum learnability from landscape fluctuation.arXiv preprint arXiv:2406.11805, 2024

    2024 arXiv

  42. [50]

    Scalable quantum archi- tecture search via landscape analysis

    Chenghong Zhu, Xian Wu, Hao-Kai Zhang, Sixuan Wu, Guangxi Li, and Xin Wang. Scalable quantum archi- tecture search via landscape analysis. arXiv preprint arXiv:2505.05380, 2025

    2025 arXiv

  43. [51]

    The quan- tum schur transform: I

    Dave Bacon, Isaac L Chuang, and Aram W Harrow. The quan- tum schur transform: I. efficient qudit circuits. arXiv preprint quant-ph/0601001, 2005

    2005 arXiv

  44. [52]

    Effi- cient quantum circuits for schur and clebsch-gordan transforms

    Dave Bacon, Isaac L Chuang, and Aram W Harrow. Effi- cient quantum circuits for schur and clebsch-gordan transforms. Physical review letters, 97(17):170502, 2006

    2006

  45. [53]

    Lie algebraic analysis and control of quantum dynamics

    Domenico D’Alessandro. Lie algebraic analysis and control of quantum dynamics. arXiv preprint arXiv:0803.1193, 2008

    2008 arXiv

  46. [54]

    QuAIRKit

    QuAIR team. QuAIRKit. https://github.com/ QuAIR/QuAIRKit, 2023. 10 Appendix for LCQNN: Linear Combination of Quantum Neural Networks Appendix A: Elements of Representation Theory Definition 1 (Unitary Representation) Let G be a group and H a Hilbert space. The unitary represen...

    2023

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.