pith. sign in

arxiv: 2604.15626 · v1 · submitted 2026-04-17 · 🪐 quant-ph

Bridge the Gap between Classical and Quantum Neural Networks with Residual Connections

Junxu Li This is my paper

Pith reviewed 2026-05-10 08:52 UTC · model grok-4.3

classification 🪐 quant-ph
keywords classicalquantumhqrnresidualanalogbridgeclassificationconnections
0
0 comments X

The pith

HQRN creates an exact functional match to classical residual networks on basis inputs while using quantum correlations for better performance on mixed states in digit recognition and entanglement classification.

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

The work builds a quantum version of a residual neural network, a type of deep learning model that uses skip connections to make training easier. When the input data is restricted to simple yes-no states that classical computers use, the quantum network evolves in exactly the same way as the classical one. This means any weights found by training a classical model can be turned straight into quantum gates without retraining from scratch. When the inputs are more general quantum states that contain mixtures and correlations, the quantum network can read extra information stored in the off-diagonal parts of the density matrix. These parts are invisible to classical networks. The authors test the idea on handwritten digit recognition and on telling apart truly entangled quantum pairs from separable states that have been crafted to match the same average measurements. The quantum version keeps high accuracy even against these adversarial look-alikes.

Core claim

We introduce a Hybrid Quantum Residual Network (HQRN) and establish an exact functional correspondence between its state evolution and the dynamics of classical networks with residual connections.

Load-bearing premise

That restricting inputs to the computational basis causes the HQRN to reduce exactly to its classical analog, enabling direct weight translation without additional quantum effects altering the dynamics.

Figures

Figures reproduced from arXiv: 2604.15626 by Junxu Li.

Figure 1
Figure 1. Figure 1: FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2 [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
read the original abstract

We introduce a Hybrid Quantum Residual Network (HQRN) and establish an exact functional correspondence between its state evolution and the dynamics of classical networks with residual connections. When inputs are restricted to the computational basis, the HQRN reduces to its classical analog, enabling the direct translation of optimized classical weights into quantum unitary operations, effectively inheriting the landscape benefits of classical optimization. Conversely, when processing general mixed states, the HQRN leverages off-diagonal quantum correlations to resolve features inaccessible to its classical analog. We validate this framework through digit recognition and bipartite entanglement classification. Notably, HQRN achieves high classification accuracy even for adversarial separable states that mimic the marginal measurement statistics of entangled pairs. Our results bridge the gap between classical and quantum residual learning, paving a scalable pathway for deep quantum architectures.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on the definition of the HQRN architecture and the unproven-in-abstract assertion that its evolution matches classical residuals exactly on basis states; no free parameters or external benchmarks are mentioned.

axioms (1)
  • domain assumption Quantum network state evolution can be engineered to match classical residual dynamics exactly when inputs are restricted to computational basis states.
    This is the load-bearing premise stated in the abstract but not derived or evidenced there.
invented entities (1)
  • Hybrid Quantum Residual Network (HQRN) no independent evidence
    purpose: To provide a quantum architecture that inherits classical residual learning benefits while accessing quantum correlations.
    Newly introduced model with no independent prior evidence cited.

pith-pipeline@v0.9.0 · 5416 in / 1294 out tokens · 68705 ms · 2026-05-10T08:52:50.688172+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

45 extracted references · 45 canonical work pages · 1 internal anchor

  1. [1]

    Schuld, I

    M. Schuld, I. Sinayskiy, and F. Petruccione, An in- troduction to quantum machine learning, Contem- porary Physics56, 172 (2015)

    work page 2015

  2. [2]

    Biamonte, P

    J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, and S. Lloyd, Quantum machine learn- ing, Nature549, 195 (2017). 7

    work page 2017

  3. [3]

    Huang, M

    H.-Y. Huang, M. Broughton, M. Mohseni, R. Bab- bush, S.Boixo, H.Neven,andJ.R.McClean,Power of data in quantum machine learning, Nature com- munications12, 2631 (2021)

    work page 2021

  4. [4]

    Sajjan, J

    M. Sajjan, J. Li, R. Selvarajan, S. H. Sureshbabu, S. S. Kale, R. Gupta, V. Singh, and S. Kais, Quan- tum machine learning for chemistry and physics, Chemical Society Reviews51, 6475 (2022)

    work page 2022

  5. [5]

    Cerezo, G

    M. Cerezo, G. Verdon, H.-Y. Huang, L. Cincio, and P. J. Coles, Challenges and opportunities in quan- tum machine learning, Nature computational sci- ence2, 567 (2022)

    work page 2022

  6. [6]

    Y.LeCun, Y.Bengio,andG.Hinton,Deeplearning, nature521, 436 (2015)

    work page 2015

  7. [7]

    Wu, W.-S

    Y. Wu, W.-S. Bao, S. Cao, F. Chen, M.-C. Chen, X. Chen, T.-H. Chung, H. Deng, Y. Du, D. Fan, et al.,Strongquantumcomputationaladvantageus- ing a superconducting quantum processor, Physical review letters127, 180501 (2021)

    work page 2021

  8. [8]

    A. J. Daley, I. Bloch, C. Kokail, S. Flannigan, N. Pearson, M. Troyer, and P. Zoller, Practical quantum advantage in quantum simulation, Nature 607, 667 (2022)

    work page 2022

  9. [9]

    Huang, M

    H.-Y. Huang, M. Broughton, J. Cotler, S. Chen, J. Li, M. Mohseni, H. Neven, R. Babbush, R. Kueng, J. Preskill,et al., Quantum advantage in learning from experiments, Science376, 1182 (2022)

    work page 2022

  10. [10]

    Aaronson, Read the fine print, Nature Physics 11, 291 (2015)

    S. Aaronson, Read the fine print, Nature Physics 11, 291 (2015)

    work page 2015

  11. [11]

    Ristè, M

    D. Ristè, M. P. Da Silva, C. A. Ryan, A. W. Cross, A. D. Córcoles, J. A. Smolin, J. M. Gambetta, J. M. Chow, and B. R. Johnson, Demonstration of quan- tum advantage in machine learning, npj Quantum Information3, 16 (2017)

    work page 2017

  12. [12]

    Huang, R

    H.-Y. Huang, R. Kueng, and J. Preskill, Information-theoretic bounds on quantum ad- vantage in machine learning, Physical Review Letters126, 190505 (2021)

    work page 2021

  13. [13]

    Anshu, S

    A. Anshu, S. Arunachalam, T. Kuwahara, and M. Soleimanifar, Sample-efficient learning of inter- acting quantum systems, Nature Physics17, 931 (2021)

    work page 2021

  14. [14]

    Lewis, D

    L. Lewis, D. Gilboa, and J. R. McClean, Quantum advantage for learning shallow neural networks with natural data distributions, Nature Communications (2025)

    work page 2025

  15. [15]

    Zhao and D.-L

    H. Zhao and D.-L. Deng, Entanglement-induced provable and robust quantum learning advantages, npj Quantum Information11, 127 (2025)

    work page 2025

  16. [16]

    Mind the gaps: The fraught road to quantum advantage

    J. Eisert and J. Preskill, Mind the gaps: The fraught road to quantum advantage, arXiv preprint arXiv:2510.19928 (2025)

    work page internal anchor Pith review arXiv 2025

  17. [17]

    Huang, S

    H.-Y. Huang, S. Choi, J. R. McClean, and J. Preskill, The vast world of quantum advantage, arXiv preprint arXiv:2508.05720 (2025)

    work page arXiv 2025

  18. [18]

    Schuld, A

    M. Schuld, A. Bocharov, K. M. Svore, and N. Wiebe, Circuit-centric quantum classifiers, Phys- ical Review A101, 032308 (2020)

    work page 2020

  19. [19]

    A. Mari, T. R. Bromley, J. Izaac, M. Schuld, and N. Killoran, Transfer learning in hybrid classical- quantum neural networks, Quantum4, 340 (2020)

    work page 2020

  20. [20]

    K. Beer, D. Bondarenko, T. Farrelly, T. J. Osborne, R. Salzmann, D. Scheiermann, and R. Wolf, Train- ing deep quantum neural networks, Nature commu- nications11, 808 (2020)

    work page 2020

  21. [21]

    Li, Knowledge distillation inspired variational quantum eigensolver with virtual annealing, New Journal of Physics28, 024501 (2026)

    J. Li, Knowledge distillation inspired variational quantum eigensolver with virtual annealing, New Journal of Physics28, 024501 (2026)

    work page 2026

  22. [22]

    Broughton, G

    M. Broughton, G. Verdon, T. McCourt, A. J. Martinez, J. H. Yoo, S. V. Isakov, P. Massey, R. Halavati, M. Y. Niu, A. Zlokapa,et al., Tensor- flow quantum: A software framework for quantum machine learning, arXiv preprint arXiv:2003.02989 (2020)

    work page arXiv 2003

  23. [23]

    I. Cong, S. Choi, and M. D. Lukin, Quantum convo- lutional neural networks, Nature Physics15, 1273 (2019)

    work page 2019

  24. [24]

    Ceschini, A

    A. Ceschini, A. Carbone, A. Sebastianelli, M. Panella, and B. Le Saux, On hybrid quanvolu- tional neural networks optimization, Quantum Ma- chine Intelligence7, 18 (2025)

    work page 2025

  25. [25]

    Y. Li, Z. Wang, R. Han, S. Shi, J. Li, R. Shang, H. Zheng, G. Zhong, and Y. Gu, Quantum recur- rent neural networks for sequential learning, Neural Networks166, 148 (2023)

    work page 2023

  26. [26]

    Takaki, K

    Y. Takaki, K. Mitarai, M. Negoro, K. Fujii, and M. Kitagawa, Learning temporal data with a varia- tional quantum recurrent neural network, Physical Review A103, 052414 (2021)

    work page 2021

  27. [27]

    Liang, W

    Y. Liang, W. Peng, Z.-J. Zheng, O. Silvén, and G. Zhao, A hybrid quantum–classical neural net- work with deep residual learning, Neural Networks 143, 133 (2021)

    work page 2021

  28. [28]

    J. Wen, Z. Huang, D. Cai, and L. Qian, Enhanc- ing the expressivity of quantum neural networks with residual connections, Communications Physics 7, 220 (2024)

    work page 2024

  29. [29]

    Kashif and S

    M. Kashif and S. Al-Kuwari, Resqnets: a residual approach for mitigating barren plateaus in quantum neural networks, EPJ Quantum Technology11, 1 (2024)

    work page 2024

  30. [30]

    J. R. McClean, S. Boixo, V. N. Smelyanskiy, R. Babbush, and H. Neven, Barren plateaus in quantum neural network training landscapes, Na- ture communications9, 4812 (2018)

    work page 2018

  31. [31]

    Larocca, S

    M. Larocca, S. Thanasilp, S. Wang, K. Sharma, J. Biamonte, P. J. Coles, L. Cincio, J. R. McClean, Z. Holmes, and M. Cerezo, Barren plateaus in varia- tional quantum computing, Nature Reviews Physics 7, 174 (2025)

    work page 2025

  32. [32]

    K. He, X. Zhang, S. Ren, and J. Sun, Identity map- pings in deep residual networks, inEuropean confer- ence on computer vision(Springer, 2016) pp. 630– 645

    work page 2016

  33. [33]

    K. He, X. Zhang, S. Ren, and J. Sun, Deep resid- ual learning for image recognition, inProceedings of the IEEE conference on computer vision and pat- tern recognition(2016) pp. 770–778

    work page 2016

  34. [34]

    Pandey, K

    D. Pandey, K. Niwaria, and B. Chourasia, Machine learning algorithms: a review, Mach. Learn6, 916 (2019)

    work page 2019

  35. [35]

    Cybenko, Approximation by superpositions of a sigmoidal function, Mathematics of control, signals and systems2, 303 (1989)

    G. Cybenko, Approximation by superpositions of a sigmoidal function, Mathematics of control, signals and systems2, 303 (1989)

    work page 1989

  36. [36]

    Hornik, Approximation capabilities of multi- layer feedforward networks, Neural networks4, 251 (1991)

    K. Hornik, Approximation capabilities of multi- layer feedforward networks, Neural networks4, 251 (1991). 8

    work page 1991

  37. [37]

    Lloyd, Universal quantum simulators, Science 273, 1073 (1996)

    S. Lloyd, Universal quantum simulators, Science 273, 1073 (1996)

    work page 1996

  38. [38]

    LeCun, L

    Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, Gradient-based learning applied to document recog- nition, Proceedings of the IEEE86, 2278 (2002)

    work page 2002

  39. [39]

    Deng, The mnist database of handwritten digit images for machine learning research [best of the web], IEEE signal processing magazine29, 141 (2012)

    L. Deng, The mnist database of handwritten digit images for machine learning research [best of the web], IEEE signal processing magazine29, 141 (2012)

    work page 2012

  40. [40]

    B. S. Nagy, C. Foias, H. Bercovici, and L. Kérchy, Harmonic analysis of operators on Hilbert space (Springer Science & Business Media, 2010)

    work page 2010

  41. [41]

    Quantum algorithms for supervised and unsupervised machine learning

    S. Lloyd, M. Mohseni, and P. Rebentrost, Quan- tum algorithms for supervised and unsupervised machine learning, arXiv preprint arXiv:1307.0411 (2013)

    work page Pith review arXiv 2013

  42. [42]

    Ostmeyer, Optimised trotter decompositions for classical and quantum computing, Journal of Physics A: Mathematical and Theoretical56, 285303 (2023)

    J. Ostmeyer, Optimised trotter decompositions for classical and quantum computing, Journal of Physics A: Mathematical and Theoretical56, 285303 (2023)

    work page 2023

  43. [43]

    R. F. Werner, Quantum states with einstein- podolsky-rosen correlations admitting a hidden- variable model, Physical Review A40, 4277 (1989)

    work page 1989

  44. [44]

    Li and S

    J. Li and S. Kais, Entanglement classifier in chemi- cal reactions, Science advances5, eaax5283 (2019)

    work page 2019

  45. [45]

    Both Correct

    J. Li and S. Kais, Quantum cluster algorithm for data classification, Materials Theory5, 6 (2021). 9 Supplementary Materials SI. THE OUTPUT OF SUCCESSIVE RESIDUAL BLOCKS In the Hybrid quantum residual network (HQRN) architecture, the output of thek-th quantum residual block (QRB) is obtained by mixing intermediate stateP n h(k) n |n⟩⟨n|with the inputρ (k−...

    work page 2021