pith. sign in

arxiv: 2509.13528 · v2 · pith:TAGHTH7Ynew · submitted 2025-09-16 · 🪐 quant-ph · cond-mat.dis-nn

Evaluating the Limits of QAOA Parameter Transfer at High-Rounds on Sparse Ising Models With Geometrically Local Cubic Terms

Elijah Pelofske , Marek Rams , Andreas B\"artschi , Piotr Czarnik , Paolo Braccia , Lukasz Cincio , Stephan Eidenbenz This is my paper

Pith reviewed 2026-05-18 15:28 UTC · model grok-4.3

classification 🪐 quant-ph cond-mat.dis-nn
keywords QAOAparameter transferIsing modelsheavy-hex graphsquantum optimizationNISQ hardwarecombinatorial optimization
0
0 comments X

The pith

QAOA angles transferred from small heavy-hex Ising instances improve expectation values on large unseen instances as depth increases.

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

The paper tests whether QAOA angles optimized on small versions of a certain type of optimization problem can be used directly on much bigger versions of the same problem. The problems are Ising models on heavy-hex graphs that include local three-variable terms. Even though adding more layers does not always make things better right away, the overall trend is toward solutions that get closer to the best possible answer. This was checked with computer simulations on up to 156 qubits and also run on actual quantum computers from IBM. The finding points to a way to use QAOA on large problems without having to tune the angles each time from scratch.

Core claim

The paper establishes that QAOA parameter transfer from single small instances to unseen large instances of the same model family does not always provide monotonically improving performance as a function of p, but the transferred angles exhibit a general trend of improved expectation value as the QAOA depth increases, in many cases converging close to the true ground-state energy of the 100+ qubit instances. Validation uses full statevector, PEPS, MPS, and LOWESA simulations plus direct runs on IBM superconducting processors.

What carries the argument

QAOA parameter transfer, in which a fixed set of angles optimized on one small instance is applied without re-optimization to larger instances from the same family of heavy-hex Ising models with local cubic terms.

If this is right

  • Solution quality improves continuously with added layers on IBM hardware up to p=5 on ibm_fez, p=9 on ibm_torino, and p=10 on ibm_pittsburgh.
  • Simulations show the transferred angles often reach expectation values near the true ground-state energy for instances over 100 qubits.
  • Parameter transfer supplies a non-variational route to usable QAOA angles on large problems.
  • Non-monotonic performance drops at intermediate depths do not block the overall upward trend with increasing p.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same transfer principle could be tested on other sparse graphs that preserve local geometric structure at different sizes.
  • Using the transferred angles as an initial guess for a short round of re-optimization on the large instance might yield further gains.
  • Hardware noise appears to cap the useful depth at processor-dependent values, suggesting noise-aware angle selection could extend the approach.

Load-bearing premise

The heavy-hex Ising models with geometrically local cubic terms are self-similar enough across scales that angles optimized on 16-27 qubit instances remain effective when applied to 100+ qubit instances.

What would settle it

A clear failure of the transferred angles to approach ground-state energy or systematic worsening of expectation value when applied to instances substantially larger than 156 qubits or to graphs lacking the same local cubic structure.

Figures

Figures reproduced from arXiv: 2509.13528 by Andreas B\"artschi, Elijah Pelofske, Lukasz Cincio, Marek Rams, Paolo Braccia, Piotr Czarnik, Stephan Eidenbenz.

Figure 1
Figure 1. Figure 1: FIG. 1. 16 qubit instance QAOA angles learned up to [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. 27 qubit instance QAOA angles learned up to [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Approximation ratio (y-axis) vs [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. Numerical simulation of parameter transfer of one of the 16 qubit problem instances onto the other 11 16 [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. Numerical simulation of parameter transfer using the donor source graph of an entirely negative coefficient Ising model, [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6. Numerical simulation of QAOA parameter transfer using the donor source graph of an entirely positive coefficient [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7. IBM quantum processor QAOA performance results for the 156 qubit IBM quantum computers in terms of objective [PITH_FULL_IMAGE:figures/full_fig_p011_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8. IBM quantum processor QAOA performance results in terms of objective function (Ising model) energy on the y-axis as [PITH_FULL_IMAGE:figures/full_fig_p012_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9. IBM quantum processor hardware error rates, aggregated from the vendor-provided calibration data for all QAOA [PITH_FULL_IMAGE:figures/full_fig_p013_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: FIG. 10. Validating the QAOA angle quality, trained on the set of 12 16-qubit instances, applied to the 156-qubit (left), [PITH_FULL_IMAGE:figures/full_fig_p015_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: FIG. 11. Validating the QAOA angle quality, trained on the set of 12 distinct 27-qubit Ising model instances, applied to the [PITH_FULL_IMAGE:figures/full_fig_p015_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: FIG. 12. PEPS simulation convergence analysis using four of the sets of parameter transferred QAOA angles (specifically from [PITH_FULL_IMAGE:figures/full_fig_p016_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: FIG. 13. Ground-state energy QAOA sampling rate (probability of sampling the ground state) from [PITH_FULL_IMAGE:figures/full_fig_p016_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: FIG. 14. Discrepancy between the most accurate PEPS simulation at [PITH_FULL_IMAGE:figures/full_fig_p018_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: FIG. 15. Heavy-hex defined Ising model with geometrically local cubic terms (left), corresponding [PITH_FULL_IMAGE:figures/full_fig_p020_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: FIG. 16. A comparison of LOWESA and PEPS energy and Hamiltonian terms convergence. We plot here results for [PITH_FULL_IMAGE:figures/full_fig_p020_16.png] view at source ↗
Figure 17
Figure 17. Figure 17: FIG. 17. Comparison of the three different PEPS contraction methods contraction methods, for a few selected representative [PITH_FULL_IMAGE:figures/full_fig_p021_17.png] view at source ↗
Figure 18
Figure 18. Figure 18: FIG. 18. PEPS convergence plots for 156-qubit QAOA circuits, using the other 8 trained angle sets from 16-qubit instances, [PITH_FULL_IMAGE:figures/full_fig_p022_18.png] view at source ↗
read the original abstract

The emergent practical applicability of the Quantum Approximate Optimization Algorithm (QAOA) for approximate combinatorial optimization is a subject of considerable interest. One of the primary limitations of QAOA is the task of finding a set of good parameters. Parameter transfer is a phenomenon where QAOA angles trained on problem instances that are self-similar tend to perform well for other problem instances from that similar class. This suggests a potentially highly efficient and scalable non-variational learning method for QAOA angle finding. We systematically study QAOA parameter transferability from small problems (16, 27 qubits) onto large problem instances (up to 156 qubits) for heavy-hex graph Ising models with geometrically local higher order terms using the Julia based QAOA simulation tool JuliQAOA to perform classical angle finding for up to 49 QAOA layers. Parameter transfer of the fixed angles is validated using a combination of full statevector, Projected Entangled Pair States, Matrix Product State, and LOWESA numerical simulations. We find that the QAOA parameter transfer from single instances applied to unseen problem instances does not in general provide monotonically improving performance as a function of p - there are many cases where the performance temporarily decreases as a function of p - but despite this the transferred angles have a general trend of improved expectation value as the QAOA depth increases, in many cases converging close to the true ground-state energy of the 100+ qubit instances. We also sample the hardware-compatible Ising models using the ensemble of fixed QAOA angles on several superconducting qubit IBM Quantum processors with 127, 133, and 156 qubits. We find continuous solution quality improvement of the hardware-compatible QAOA circuits run on the IBM NISQ processors up to p=5 on ibm_fez, p=9 on ibm_torino, and p=10 on ibm_pittsburgh.

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.

Referee Report

2 major / 2 minor

Summary. The manuscript empirically studies QAOA parameter transfer on heavy-hex Ising models with geometrically local cubic terms. Angles are classically optimized on small instances (16-27 qubits) and transferred without re-optimization to large unseen instances (up to 156 qubits) for depths up to p=49. Multiple backends (statevector, PEPS, MPS, LOWESA) plus IBM hardware runs are used to evaluate performance; the central observation is a general improving trend in expectation value with depth despite non-monotonicity, with many cases approaching the reported ground-state energy.

Significance. If the reference energies are accurate, the results indicate that parameter transfer from single small instances can yield scalable, non-variational QAOA angles that improve with depth on sparse, geometrically structured models, supporting practical use on NISQ hardware without per-instance optimization. The multi-method cross-validation and hardware data up to p=10 provide concrete empirical support for the trend.

major comments (2)
  1. [Numerical Simulations and Results] Numerical validation sections: the bond-dimension truncation errors for MPS, PEPS, and LOWESA are not quantified or shown to converge for p up to 49 on 100+ qubit heavy-hex graphs with cubic terms. High-depth QAOA states generically develop volume-law entanglement, so fixed-bond approximations can produce systematically biased energies; without controlled error estimates, claims that transferred angles converge close to the true ground-state energy rest on unverified reference values.
  2. [Large Instance Evaluation] Large-instance results: the reported closeness to ground-state energy on 100+ qubit instances is load-bearing for the central claim, yet the paper provides no independent verification (e.g., exact diagonalization on smaller proxies or extrapolation of truncation error) that the tensor-network energies are accurate to the precision needed to support the convergence statement.
minor comments (2)
  1. [Figures] Figure captions and axis labels should explicitly state the bond dimensions or truncation thresholds used for each backend at each p to allow readers to assess approximation quality.
  2. [Abstract and Results] The abstract states 'converging close to the true ground-state energy' while the text notes non-monotonicity; a brief quantitative statement of the typical deviation from the reported GS energy at highest p would improve clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful and constructive review of our manuscript. The comments on numerical validation and reference energy accuracy are well-taken, and we outline targeted revisions below to address them while preserving the core empirical findings on parameter transfer.

read point-by-point responses

  1. Referee: [Numerical Simulations and Results] Numerical validation sections: the bond-dimension truncation errors for MPS, PEPS, and LOWESA are not quantified or shown to converge for p up to 49 on 100+ qubit heavy-hex graphs with cubic terms. High-depth QAOA states generically develop volume-law entanglement, so fixed-bond approximations can produce systematically biased energies; without controlled error estimates, claims that transferred angles converge close to the true ground-state energy rest on unverified reference values.

    Authors: We agree that explicit quantification of truncation errors strengthens the claims. In the revised manuscript we will add a dedicated appendix with bond-dimension scaling studies on representative 100+ qubit instances at selected high p values (including p=49), reporting energy differences as a function of bond dimension for MPS, PEPS, and LOWESA. We will also include a brief discussion of entanglement growth on these sparse, geometrically local models and why the observed cross-method agreement supports the reported trends despite possible volume-law contributions. revision: yes

  2. Referee: [Large Instance Evaluation] Large-instance results: the reported closeness to ground-state energy on 100+ qubit instances is load-bearing for the central claim, yet the paper provides no independent verification (e.g., exact diagonalization on smaller proxies or extrapolation of truncation error) that the tensor-network energies are accurate to the precision needed to support the convergence statement.

    Authors: We acknowledge the need for additional verification. The revision will incorporate (i) exact-diagonalization comparisons on smaller proxy heavy-hex instances with cubic terms to benchmark the tensor-network methods, and (ii) explicit extrapolation of truncation error versus bond dimension for the largest instances. We note that the hardware results (up to p=10 on 127–156 qubit devices) already provide an independent, non-tensor-network confirmation of the improving trend with depth. revision: yes

Circularity Check

0 steps flagged

Empirical numerical study exhibits no circular derivation chain

full rationale

The paper reports direct numerical experiments: QAOA angles are classically optimized on small (16-27 qubit) heavy-hex Ising instances and then transferred without re-optimization to larger (up to 156 qubit) instances of the same family. Expectation values are evaluated via statevector, MPS, PEPS, and LOWESA simulations on the target instances. No first-principles derivation, uniqueness theorem, or ansatz is invoked whose validity reduces to the fitted angles or to self-citation of the present work. The central observations (non-monotonic but generally improving performance with depth, occasional convergence toward reference energies) are statistical outcomes of these independent computations rather than algebraic identities or re-labeled fits. Any self-citations present are incidental and non-load-bearing for the empirical claims.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The study relies on standard quantum mechanics and classical optimization of QAOA angles; no new physical axioms or invented entities are introduced. Free parameters are the QAOA angles themselves, which are fitted per small instance.

free parameters (1)
  • QAOA angles (gamma, beta) per layer
    Angles are variationally optimized on small instances using classical methods and then held fixed for transfer.
axioms (1)
  • standard math Standard Schrödinger evolution and measurement postulates for QAOA circuit simulation
    Invoked implicitly when using statevector, PEPS, and MPS simulators to compute expectation values.

pith-pipeline@v0.9.0 · 5900 in / 1443 out tokens · 32828 ms · 2026-05-18T15:28:13.522536+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 1 Pith paper

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

  1. Tensor network surrogate models for variational quantum computation

    quant-ph 2026-04 unverdicted novelty 6.0

    Tensor network simulations act as effective surrogate models for training QAOA on large 2D lattices, overcoming limits of parameter transfer from small instances and remaining classically feasible with moderate bond d...

Reference graph

Works this paper leans on

103 extracted references · 103 canonical work pages · cited by 1 Pith paper · 21 internal anchors

  1. [1]

    A Quantum Approximate Optimization Algorithm

    E. Farhi, J. Goldstone, and S. Gutmann, A Quantum Approximate Optimization Algorithm, arXiv preprint (2014), arXiv:1411.4028

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  2. [2]

    A Quantum Approximate Optimization Algorithm Applied to a Bounded Occurrence Constraint Problem

    E. Farhi, J. Goldstone, and S. Gutmann, A Quan- tum Approximate Optimization Algorithm Applied to a Bounded Occurrence Constraint Problem, arXiv preprint (2015), arXiv:1412.6062

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  3. [3]

    From the Quantum Approximate Optimization Algorithm to a Quantum Alternating Operator Ansatz

    S. Hadfield, Z. Wang, B. O 'Gorman, E. Rieffel, D. Ven- turelli, and R. Biswas, From the Quantum Approx- imate Optimization Algorithm to a Quantum Alter- nating Operator Ansatz, Algorithms 12, 34 (2019), arXiv:1709.03489

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  4. [4]

    The Quantum Approximate Optimization Algorithm Needs to See the Whole Graph: A Typical Case, April 2020

    E. Farhi, D. Gamarnik, and S. Gutmann, The Quantum Approximate Optimization Algorithm Needs to See the Whole Graph: A Typical Case, arXiv preprint (2020), arXiv:2004.09002

    work page arXiv 2020

  5. [5]

    Farhi, D

    E. Farhi, D. Gamarnik, and S. Gutmann, The Quantum Approximate Optimization Algorithm Needs to See the Whole Graph: Worst Case Examples, arXiv preprint (2020), arXiv:2005.08747

    work page arXiv 2020

  6. [6]

    Quantum Computing in the NISQ era and beyond

    J. Preskill, Quantum Computing in the NISQ era and beyond, Quantum 2, 79 (2018), arXiv:1801.00862

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  7. [7]

    Quantum supremacy through the quantum approx- imate optimization algorithm

    E. Farhi and A. W. Harrow, Quantum Supremacy through the Quantum Approximate Optimization Al- gorithm, arXiv preprint (2019), arXiv:1602.07674

    work page arXiv 2019

  8. [8]

    Evidence of scaling advantage for the quantum approxi- mate optimization algorithm on a classically intractable problem

    R. Shaydulin, C. Li, S. Chakrabarti, M. DeCross, D. Herman, N. Kumar, J. Larson, D. Lykov, P. Minssen, Y. Sun, Y. Alexeev, J. M. Dreiling, J. P. Gaebler, T. M. Gatterman, J. A. Gerber, K. Gilmore, D. Gresh, N. Hewitt, C. V. Horst, S. Hu, J. Johansen, M. Ma- theny, T. Mengle, M. Mills, S. A. Moses, B. Neyen- huis, P. Siegfried, R. Yalovetzky, and M. Pistoi...

    work page arXiv 2024

  9. [9]

    M. P. Harrigan, K. J. Sung, M. Neeley, K. J. Satzinger, F. Arute, K. Arya, J. Atalaya, J. C. Bardin, R. Barends, S. Boixo, M. Broughton, B. B. Buck- ley, D. A. Buell, B. Burkett, N. Bushnell, Y. Chen, Z. Chen, B. Chiaro, R. Collins, W. Courtney, S. De- mura, A. Dunsworth, D. Eppens, A. Fowler, B. Foxen, C. Gidney, M. Giustina, R. Graff, S. Habegger, A. Ho...

    work page arXiv 2021

  10. [10]

    Pelofske, A

    E. Pelofske, A. B¨ artschi, J. Golden, and S. Eidenbenz, High-Round QAOA for MAX k-SAT on Trapped Ion NISQ Devices, in IEEE International Conference on Quantum Computing and Engineering QCE’23 (2023) p. 506–517, arXiv:2306.03238

    work page arXiv 2023

  11. [11]

    Shaydulin and M

    R. Shaydulin and M. Pistoia, QAOA with N · p ≥ 200, in IEEE International Conference on Quantum Com- puting and Engineering QCE’23 (2023) p. 1074–1077, arXiv:2303.02064

    work page arXiv 2023

  12. [12]

    Weidenfeller, L

    J. Weidenfeller, L. C. Valor, J. Gacon, C. Tornow, L. Bello, S. Woerner, and D. J. Egger, Scaling of the quantum approximate optimization algorithm on su- perconducting qubit based hardware, Quantum 6, 870 (2022), arXiv:2202.03459

    work page arXiv 2022

  13. [13]

    Golden, A

    J. Golden, A. B¨ artschi, S. Eidenbenz, and D. O’Malley, Numerical Evidence for Exponential Speed-up of QAOA over Unstructured Search for Approximate Constrained Optimization, in IEEE International Conference on Quantum Computing and Engineering QCE’23 (2023) p. 496–505, arXiv:2202.00648

    work page arXiv 2023

  14. [14]

    Boulebnane and A

    S. Boulebnane and A. Montanaro, Solving Boolean Sat- isfiability Problems With The Quantum Approximate Optimization Algorithm, PRX Quantum 5, 030348 (2024), arXiv:2208.06909

    work page arXiv 2024

  15. [15]

    Variational quantum algorithms

    M. Cerezo, A. Arrasmith, R. Babbush, S. C. Benjamin, S. Endo, K. Fujii, J. R. McClean, K. Mitarai, X. Yuan, L. Cincio, and P. J. Coles, Variational quantum al- gorithms, Nature Reviews Physics 3, 625–644 (2021), arXiv:2012.09265

    work page arXiv 2021

  16. [16]

    The QAOA parameter similarity property comes by several different terms, including parameter transfer , problem instance independent QAOA parameters , fixed angles, parameter concentration, parameter fixing, size invariant QAOA parameters , fixed control parameters, and transfer learning

    work page

  17. [17]

    Akshay, D

    V. Akshay, D. Rabinovich, E. Campos, and J. Bia- monte, Parameter Concentration in Quantum Approx- imate Optimization, Physical Review A 104, L010401 (2021), arXiv:2103.11976

    work page arXiv 2021

  18. [18]

    X. Lee, Y. Saito, D. Cai, and N. Asai, Parameters Fixing Strategy for Quantum Approximate Optimization Algo- rithm, in IEEE International Conference on Quantum Computing and Engineering QCE’21 (2021) p. 10–16, arXiv:2108.05288

    work page arXiv 2021

  19. [19]

    The quantum approximate op- timization algorithm at high depth for maxcut on large- girth regular graphs and the sherrington-kirkpatrick model

    J. Basso, E. Farhi, K. Marwaha, B. Villalonga, and L. Zhou, The Quantum Approximate Optimization Al- gorithm at High Depth for MaxCut on Large-Girth Reg- ular Graphs and the Sherrington-Kirkpatrick Model, in 17th Conference on the Theory of Quantum Computa- tion, Communication and Cryptography TQC’22 (2022) arXiv:2110.14206

    work page arXiv 2022

  20. [20]

    The Quantum Approximate Optimization Algorithm and the Sherrington-Kirkpatrick Model at Infinite Size.Quantum, 6: 759, July 2022

    E. Farhi, J. Goldstone, S. Gutmann, and L. Zhou, The Quantum Approximate Optimization Algorithm and the Sherrington-Kirkpatrick Model at Infinite Size, Quantum 6, 759 (2022), arXiv:1910.08187

    work page arXiv 2022

  21. [21]

    & Montanaro, A

    S. Boulebnane and A. Montanaro, Predicting param- eters for the quantum approximate optimization algo- rithm for max-cut from the infinite-size limit, arXiv preprint (2021), arXiv:2110.10685

    work page arXiv 2021

  22. [22]

    Shaydulin, P

    R. Shaydulin, P. C. Lotshaw, J. Larson, J. Ostrowski, and T. S. Humble, Parameter Transfer for Quantum Approximate Optimization of Weighted MaxCut, ACM Transactions on Quantum Computing 4, 19 (2023), 24 arXiv:2201.11785

    work page arXiv 2023

  23. [23]

    S. H. Sureshbabu, D. Herman, R. Shaydulin, J. Basso, S. Chakrabarti, Y. Sun, and M. Pistoia, Parame- ter Setting in Quantum Approximate Optimization of Weighted Problems, Quantum 8, 1231 (2024), arXiv:2305.15201

    work page arXiv 2024

  24. [24]

    The fixed angle conjecture for QAOA on regular MaxCut graphs.arXiv preprint arXiv:2107.00677, 2021

    J. Wurtz and D. Lykov, Fixed-angle conjectures for the quantum approximate optimization algorithm on reg- ular MaxCut graphs, Physical Review A 104, 052419 (2021), arXiv:2107.00677

    work page arXiv 2021

  25. [25]

    Wurtz and P

    J. Wurtz and P. Love, Maxcut quantum approxi- mate optimization algorithm performance guarantees for p > 1, Physical Review A 103, 042612 (2021), arXiv:2010.11209

    work page arXiv 2021

  26. [26]

    Streif and M

    M. Streif and M. Leib, Training the Quantum Approxi- mate Optimization Algorithm without access to a Quan- tum Processing Unit, Quantum Science and Technology 5, 034008 (2020), arXiv:1908.08862

    work page arXiv 2020

  27. [27]

    J. A. Monta˜ nez-Barrera, D. Willsch, and K. Michielsen, Transfer learning of optimal QAOA parameters in com- binatorial optimization, Quantum Information Process- ing 24, 129 (2025), arXiv:2402.05549

    work page arXiv 2025

  28. [28]

    Pelofske, A

    E. Pelofske, A. B¨ artschi, L. Cincio, J. Golden, and S. Ei- denbenz, Scaling whole-chip QAOA for higher-order ising spin glass models on heavy-hex graphs, npj Quan- tum Information 10, 109 (2024), arXiv:2312.00997

    work page arXiv 2024

  29. [29]

    F. G. S. L. Brandao, M. Broughton, E. Farhi, S. Gut- mann, and H. Neven, For fixed control parameters the quantum approximate optimization algorithm’s objec- tive function value concentrates for typical instances, arXiv preprint (2018), arXiv:1812.04170

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  30. [30]

    Sakai, H

    R. Sakai, H. Matsuyama, W.-H. Tam, Y. Ya- mashiro, and K. Fujii, Linearly simplified QAOA pa- rameters and transferability, arXiv preprint (2024), arXiv:2405.00655

    work page arXiv 2024

  31. [31]

    Katial, K

    V. Katial, K. Smith-Miles, C. Hill, and L. Hollen- berg, On the Instance Dependence of Optimal Param- eters for the Quantum Approximate Optimisation Al- gorithm: Insights via Instance Space Analysis, IN- FORMS Journal on Computing 37, 146–171 (2025), arXiv:2401.08142

    work page arXiv 2025

  32. [32]

    J. Sud, S. Hadfield, E. Rieffel, N. Tubman, and T. Hogg, Parameter-setting heuristic for the quantum alternating operator ansatz, Physical Review Research 6, 023171 (2024), arXiv:2211.09270

    work page arXiv 2024

  33. [33]

    Ozaeta, W

    A. Ozaeta, W. van Dam, and P. L. McMahon, Ex- pectation values from the single-layer quantum ap- proximate optimization algorithm on Ising problems, Quantum Science and Technology 7, 045036 (2022), arXiv:2012.03421

    work page arXiv 2022

  34. [34]

    Z. He, R. Shaydulin, D. Herman, C. Li, R. Raymond, S. H. Sureshbabu, and M. Pistoia, Parameter Setting Heuristics Make the Quantum Approximate Optimiza- tion Algorithm Suitable for the Early Fault-Tolerant Era, in 43rd IEEE/ACM International Conference on Computer-Aided Design ICCAD’24 (2024) p. 63:1–36:7, arXiv:2408.09538

    work page arXiv 2024

  35. [35]

    S. H. Sack and M. Serbyn, Quantum annealing initial- ization of the quantum approximate optimization algo- rithm, Quantum 5, 491 (2021), arXiv:2101.05742

    work page arXiv 2021

  36. [36]

    J. A. Monta˜ nez-Barrera and K. Michielsen, Toward a linear-ramp QAOA protocol: evidence of a scaling advantage in solving some combinatorial optimization problems, npj Quantum Information 11, 131 (2025), arXiv:2405.09169

    work page arXiv 2025

  37. [37]

    Pelofske, A

    E. Pelofske, A. B¨ artschi, and S. Eidenbenz, Short-Depth QAOA Circuits and Quantum Annealing on Higher- Order Ising Models, npj Quantum Information 10, 30 (2024)

    work page 2024

  38. [38]

    Pelofske, A

    E. Pelofske, A. B¨ artschi, and S. Eidenbenz, Quantum Annealing vs. QAOA: 127 Qubit Higher-Order Ising Problems on NISQ Computers, in International Con- ference on High Performance Computing ISC HPC’23 (2023) p. 240–258, arXiv:2301.00520

    work page arXiv 2023

  39. [39]

    S. V. Romero, A.-M. Visuri, A. G. Cadavid, E. Solano, and N. N. Hegade, Bias-field digitized counterdia- batic quantum algorithm for higher-order binary op- timization, Communications Physics 8, 348 (2025), arXiv:2409.04477

    work page arXiv 2025

  40. [40]

    S. H. Sack and D. J. Egger, Large-scale quantum ap- proximate optimization on nonplanar graphs with ma- chine learning noise mitigation, Physical Review Re- search 6, 013223 (2024), arXiv:2307.14427

    work page arXiv 2024

  41. [41]

    The QAOA parameter p can be found under several dif- ferent terms in the literature, including layers, rounds, depth (importantly not the same quantity as circuit depth), and levels, in this study we will typically refer to this parameter as either rounds or depth

    work page

  42. [42]

    Julia: A Fresh Approach to Numerical Computing

    J. Bezanson, A. Edelman, S. Karpinski, and V. B. Shah, Julia: A Fresh Approach to Numerical Computing, SIAM review 59, 65–98 (2017), arXiv:1411.1607

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  43. [43]

    Golden, A

    J. Golden, A. Baertschi, D. O’Malley, E. Pelofske, and S. Eidenbenz, JuliQAOA: Fast, Flexible QAOA Sim- ulation, in Workshops of The International Confer- ence on High Performance Computing, Network, Stor- age, and Analysis SC-W’23 (ACM, 2023) p. 1454–1459, arXiv:2312.06451

    work page arXiv 2023

  44. [44]

    Golden, A

    J. Golden, A. B¨ artschi, D. O’Malley, and S. Ei- denbenz, Threshold-Based Quantum Optimization, in IEEE International Conference on Quantum Comput- ing and Engineering QCE’21 (IEEE, 2021) p. 137–147, arXiv:2106.13860

    work page arXiv 2021

  45. [45]

    Golden, A

    J. Golden, A. B¨ artschi, D. O’Malley, and S. Eidenbenz, The Quantum Alternating Operator Ansatz for Satisfi- ability Problems, in IEEE International Conference on Quantum Computing and Engineering QCE’23 (IEEE,

    work page

  46. [46]

    307–312, arXiv:2301.11292

    p. 307–312, arXiv:2301.11292

    work page arXiv

  47. [47]

    Pagano, A

    G. Pagano, A. Bapat, P. Becker, K. S. Collins, A. De, P. W. Hess, H. B. Kaplan, A. Kyprianidis, W. L. Tan, C. Baldwin, L. T. Brady, A. Deshpande, F. Liu, S. Jor- dan, A. V. Gorshkov, and C. Monroe, Quantum approx- imate optimization of the long-range ising model with a trapped-ion quantum simulator, Proceedings of the Na- tional Academy of Sciences 117, 2...

    work page arXiv 2020

  48. [48]

    S. H. Sack, R. A. Medina, R. Kueng, and M. Ser- byn, Recursive greedy initialization of the quantum approximate optimization algorithm with guaranteed improvement, Physical Review A 107, 062404 (2023), arXiv:2209.01159

    work page arXiv 2023

  49. [49]

    A. Apte, S. H. Sureshbabu, R. Shaydulin, S. Bouleb- nane, Z. He, D. Herman, J. Sud, and M. Pistoia, It- erative interpolation schedules for quantum approxi- mate optimization algorithm, arXiv preprint (2025), arXiv:2504.01694

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  50. [50]

    Chamberland, G

    C. Chamberland, G. Zhu, T. J. Yoder, J. B. Hertzberg, and A. W. Cross, Topological and Subsystem Codes on Low-Degree Graphs with Flag Qubits, Physical Review 25 X 10, 011022 (2020), arXiv:1907.09528

    work page arXiv 2020

  51. [51]

    & Bermudez, A

    C. Benito, E. L´ opez, B. Peropadre, and A. Bermudez, Comparative study of quantum error correction strate- gies for the heavy-hexagonal lattice, Quantum 9, 1623 (2025), arXiv:2402.02185

    work page arXiv 2025

  52. [52]

    & Wootton, J

    B. Het´ enyi and J. R. Wootton, Creating Entan- gled Logical Qubits in the Heavy-Hex Lattice with Topological Codes, PRX Quantum 5, 040334 (2024), arXiv:2404.15989

    work page arXiv 2024

  53. [53]

    S. V. Barron, D. J. Egger, E. Pelofske, A. B¨ artschi, S. Eidenbenz, M. Lehmkuehler, and S. Woerner, Prov- able bounds for noise-free expectation values computed from noisy samples, Nature Computational Science 4, 865–875 (2024), arXiv:2312.00733

    work page arXiv 2024

  54. [54]

    Ef- ficient online quantum circuit learning with no up- front training,

    T. O’Leary, P. Czarnik, E. Pelofske, A. T. Sornborger, M. McKerns, and L. Cincio, Efficient Online Quan- tum Circuit Learning with No Upfront Training, arXiv preprint (2025), arXiv:2501.04636

    work page arXiv 2025

  55. [55]

    Classical Combinatorial Optimization Scaling for Random Ising Models on 2D Heavy-Hex Graphs

    E. Pelofske, A. B¨ artschi, and S. Eidenbenz, Classi- cal Combinatorial Optimization Scaling for Random Ising Models on 2D Heavy-Hex Graphs, arXiv preprint (2024), arXiv:2412.15572

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  56. [56]

    A. A. Hagberg, D. A. Schult, and P. J. Swart, Exploring Network Structure, Dynamics, and Function using Net- workX, in 7th Python in Science Conference SciPy’08 (2008) p. 11–15

    work page 2008

  57. [57]

    Kotil, E

    A. Kotil, E. Pelofske, S. Riedm¨ uller, D. J. Egger, S. Ei- denbenz, T. Koch, and S. Woerner, Quantum Approx- imate Multi-Objective Optimization, arXiv preprint (2025), arXiv:2503.22797

    work page arXiv 2025

  58. [58]

    J. M. Chow, A. D. C´ orcoles, J. M. Gambetta, C. Rigetti, B. R. Johnson, J. A. Smolin, J. R. Rozen, G. A. Keefe, M. B. Rothwell, M. B. Ketchen, and M. Steffen, Sim- ple All-Microwave Entangling Gate for Fixed-Frequency Superconducting Qubits, Physical Review Letters 107, 080502 (2011), arXiv:1106.0553

    work page internal anchor Pith review Pith/arXiv arXiv 2011

  59. [59]

    Qiskit contributors, Qiskit: An Open-source Framework for Quantum Computing (2023)

    work page 2023

  60. [60]

    Quantum computing with Qiskit

    A. Javadi-Abhari, M. Treinish, K. Krsulich, C. J. Wood, J. Lishman, J. Gacon, S. Martiel, P. D. Nation, L. S. Bishop, A. W. Cross, B. R. Johnson, and J. M. Gam- betta, Quantum computing with Qiskit, arXiv preprint (2024), arXiv:2405.08810

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  61. [61]

    IBM ILOG CPLEX, V12.10.0 : User’s Manual for CPLEX, International Business Machines Corporation 46, 157 (2019)

    work page 2019

  62. [62]

    Gurobi Optimization, LLC, Gurobi Optimizer Refer- ence Manual (2024)

    work page 2024

  63. [63]

    Efficient classical simulation of slightly entangled quantum computations

    G. Vidal, Efficient Classical Simulation of Slightly En- tangled Quantum Computations, Physical Review Let- ters 91, 147902 (2003), arXiv:quant-ph/0301063

    work page internal anchor Pith review Pith/arXiv arXiv 2003

  64. [64]

    Classical simulation of quantum many-body systems with a tree tensor network

    Y.-Y. Shi, L.-M. Duan, and G. Vidal, Classical sim- ulation of quantum many-body systems with a tree tensor network, Physical Review A 74, 022320 (2006), arXiv:quant-ph/0511070

    work page internal anchor Pith review Pith/arXiv arXiv 2006

  65. [65]

    On the simulation of quantum circuits

    R. Jozsa, On the simulation of quantum circuits, arXiv preprint (2006), arXiv:quant-ph/0603163

    work page internal anchor Pith review Pith/arXiv arXiv 2006

  66. [66]

    Alkabetz and I

    R. Alkabetz and I. Arad, Tensor networks contraction and the belief propagation algorithm, Physical Review Research 3, 023073 (2021), arXiv:2008.04433

    work page arXiv 2021

  67. [67]

    Tindall and M

    J. Tindall and M. Fishman, Gauging tensor networks with belief propagation, SciPost Physics 15, 222 (2023), arXiv:2306.17837

    work page arXiv 2023

  68. [68]

    H. C. Jiang, Z. Y. Weng, and T. Xiang, Accurate De- termination of Tensor Network State of Quantum Lat- tice Models in Two Dimensions, Physical Review Let- ters 101, 090603 (2008), arXiv:0806.3719

    work page internal anchor Pith review Pith/arXiv arXiv 2008

  69. [69]

    Tindall, M

    J. Tindall, M. Fishman, E. M. Stoudenmire, and D. Sels, Efficient Tensor Network Simulation of IBM’s Eagle Kicked Ising Experiment, PRX Quantum 5, 010308 (2024), arXiv:2306.14887

    work page arXiv 2024

  70. [70]

    Beguˇ si´ c, J

    T. Beguˇ si´ c, J. Gray, and G. K.-L. Chan, Fast and con- verged classical simulations of evidence for the utility of quantum computing before fault tolerance, Science Advances 10, eadk4321 (2024), arXiv:2308.05077

    work page arXiv 2024

  71. [71]

    Patra, S

    S. Patra, S. S. Jahromi, S. Singh, and R. Or´ us, Efficient tensor network simulation of IBM’s largest quantum processors, Physical Review Research 6, 013326 (2024), arXiv:2309.15642

    work page arXiv 2024

  72. [72]

    M. S. Rudolph and J. Tindall, Simulating and Sampling from Quantum Circuits with 2D Tensor Networks, arXiv preprint (2025), arXiv:2507.11424

    work page arXiv 2025

  73. [73]

    M. M. Rams, G. W´ ojtowicz, A. Sinha, and J. Hasik, YASTN: Yet another symmetric tensor networks; A Python library for Abelian symmetric tensor network calculations, SciPost Physics Codebases , 52 (2025), arXiv:2405.12196

    work page arXiv 2025

  74. [74]

    Fishman, S

    M. Fishman, S. R. White, and E. M. Stoudenmire, The ITensor Software Library for Tensor Network Calculations, SciPost Physics Codebases , 4 (2022), arXiv:2007.14822

    work page arXiv 2022

  75. [75]

    Fishman, S

    M. Fishman, S. R. White, and E. M. Stoudenmire, Codebase release 0.3 for ITensor, SciPost Physics Code- bases , 4 (2022)

    work page 2022

  76. [76]

    Rams, PEPS Heavy-Hex QAOA Simulation Github (2025)

    M. Rams, PEPS Heavy-Hex QAOA Simulation Github (2025)

    work page 2025

  77. [77]

    Fontana, M

    E. Fontana, M. S. Rudolph, R. Duncan, I. Rungger, and C. Cˆ ırstoiu, Classical simulations of noisy varia- tional quantum circuits, npj Quantum Information 11, 84 (2025), arXiv:2306.05400

    work page arXiv 2025

  78. [78]

    M. S. Rudolph, E. Fontana, Z. Holmes, and L. Cincio, Classical surrogate simulation of quantum systems with LOWESA, arXiv preprint (2023), arXiv:2308.09109

    work page arXiv 2023

  79. [79]

    Vidal, Efficient classical simulation of slightly entan- gled quantum computations, Physical review letters 91, 147902 (2003)

    G. Vidal, Efficient classical simulation of slightly entan- gled quantum computations, Physical review letters 91, 147902 (2003)

    work page 2003

  80. [80]

    Or´ us, A practical introduction to tensor networks: Matrix product states and projected entangled pair states, Annals of physics 349, 117 (2014)

    R. Or´ us, A practical introduction to tensor networks: Matrix product states and projected entangled pair states, Annals of physics 349, 117 (2014)

    work page 2014

Showing first 80 references.