Pith. sign in

REVIEW 3 major objections 6 minor 47 references

An Efficient Iterative Algorithm for Qubit Mapping via Layer-Weight Assignment and Search Space Reduction

T0 review · 3 major / 6 minor · reviewed 2026-08-08 · deepseek-v4-flash

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

Pith's one-line read Qubit-mapping heuristic claims 20.62 percent fewer SWAP gates than the previous best method.

desk verdict Plausible qubit-mapping heuristic with a clear method but an unverifiable headline comparison; worth reviewing if baseline data and code are supplied. read the letter →

arxiv 2502.07536 v2 pith:UBTMOLZ3 submitted 2025-02-11 quant-ph cs.ET

Kang Xu , Zeyang Li , Xinjian Liu , Dandan Li , Yukun Wang This is my paper · ORCID
classification quant-phcs.ET
keywords qubitmappingSWAPgateminimizationheuristicalgorithmsubgraphisomorphismlayer-weightassignmentiterativeoptimizationquantumcircuitcompilationNISQ
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

A quantum computer can execute two-qubit gates only between physically adjacent qubits, so a compiler must insert SWAP gates to move qubits; each extra gate adds time and error. This paper proposes a heuristic, HAIL, designed to insert far fewer of those gates than current methods. HAIL weights circuit layers so that early, high-parallelism gates guide the initial qubit placement, then searches over short SWAP sequences and refines the result with a few forward-backward passes. The paper reports concrete reductions on standard benchmarks, and a faster variant, HAIL-imp, that trades a small amount of quality for a large reduction in runtime. If the reported results hold, the practical payoff is cheaper and less error-prone execution of near-term quantum circuits on today's hardware.

What carries the argument

The load-bearing mechanism is the layer-weight assignment: each gate gets weight $W_g(g_i)=(\mathrm{Depth}(LC)-\mathrm{Layer}(g_i))+1$, and each edge of the qubit interaction graph accumulates those weights. The VF2 subgraph-isomorphism algorithm then embeds the heaviest edges into the device's coupling graph, and a completion step places the remaining logical qubits by maximizing $F_{val}(q,v)=\sum_u (\mathrm{dia}(AG)-\mathrm{dist}(u,v))E_w(u,v)$. Routing evaluates candidate SWAP sequences of depth $d$ using $G_{val}(\tau_{current},seq_i)$, the number of CNOT gates executable after applying the sequence divided by the sequence length, and breaks ties with a look-ahead scoring function based on gate weights and distances. Finally, a simplified iterative framework alternates forward and backward traversals a few times and keeps the configuration with the fewest inserted SWAP gates.

What would settle it

Run the published SABRE, ILS, and TWP codes on the same 23 B23 circuits on the IBM Q20 coupling map, with identical SWAP/cancellation conventions and multiple random seeds, and check whether HAIL-3's total inserted-CNOT count remains 20.62% below TWP. The paper itself reports a -2.48% case for z4_268 versus TWP, so a uniform rerun would show whether the aggregate improvement is robust or driven by favorable experimental conditions.

Watch

Extended reading notes

Core claim

The paper's central claim is that HAIL reduces the number of additional SWAP gates a compiler must insert when mapping a logical circuit onto a fixed hardware coupling graph. On the 23-circuit benchmark B23 over the IBM Q20 architecture, the depth-3 version HAIL-3 inserts 13,047 additional CNOT gates, against 16,437 for TWP, 17,811 for ILS, and 50,874 for SABRE, which the paper summarizes as a 20.62% reduction over the strongest comparator TWP. The paper further claims that the partially extended variant HAIL-imp cuts runtime by roughly 90% relative to HAIL-3 while inserting only about 5% more CNOT gates, and that on the Google Sycamore architecture it reduces inserted CNOTs on one benchmark by 37.8% and runtime by 71.52% compared with TWP. The paper's own comparison table also reports a few small circuits where HAIL-3 ties or loses to a comparator, such as z4_268 versus TWP at -2.48%, which the authors attribute to local optima in heuristic search.

Load-bearing premise

The comparison assumes that the reported SABRE, ILS, and TWP gate counts were produced under the same benchmark circuits, coupling graphs, and evaluation conventions as HAIL's runs; the paper does not state how those baseline numbers were obtained, so if they came from different setups the 20.62% improvement over TWP could be an artifact of the comparison rather than of the algorithm.

Editorial extensions

If this is right

  • If the reported B23 totals are correct, HAIL-3 would reduce inserted CNOT gates by 74.35% relative to SABRE, 26.75% relative to ILS, and 20.62% relative to TWP on that benchmark.
  • HAIL-imp gives a practical runtime-quality balance: roughly a 90% runtime reduction over HAIL-3 with only about a 5% increase in inserted CNOT gates.
  • On the Google Sycamore architecture, the paper's experiments show HAIL-imp reducing inserted CNOTs by about 37.8% and runtime by 71.52% on one benchmark relative to TWP, with at least 50% runtime savings on three of four benchmarks.
  • The paper's complexity analysis implies that search depth $d$ dominates the cost, since the candidate set scales as $O(|E|^d)$, which is why the depth-3 variant is presented as the recommended operating point.

Reading between the lines

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

  • Editorial inference: if the baseline numbers reproduce, the layer-weight initial mapping is separable from the rest of HAIL and could likely be dropped into other routing heuristics to improve their starting points as well.
  • Editorial inference: the extend-top-K-depth-2-sequences idea is a general way to tame exponential search blowup in combinatorial search problems, not just in qubit routing; the same partial-extension trick could apply to other look-ahead searches.
  • Editorial inference: replacing the circuit-depth weight in Eq. (1) with an error- or noise-aware weight would turn HAIL from minimizing gate count into minimizing expected execution error, a direction the paper does not explore.
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 / 6 minor

Summary. The paper proposes HAIL, an iterative heuristic for qubit mapping on near-term hardware. The method combines a layer-weight-based initial mapping built from subgraph isomorphism and greedy completion, a two-stage SWAP-sequence search that prunes candidates and then refines ties, and a forward-backward iterative refinement framework. A variant, HAIL-imp, reduces the search space by extending only the most promising depth-2 sequences to depth 3. The paper reports on the IBM Q20 and Google Sycamore architectures that HAIL and HAIL-imp insert fewer CNOT/SWAP gates than SABRE, ILS, and TWP, with the headline claim of a 20.62% reduction over TWP on the B23 benchmark and a 90% runtime reduction for HAIL-imp relative to HAIL-3. The evidence is entirely experimental; no optimality guarantee is proved.

Significance. If the experimental comparison is controlled and reproducible, the paper would be a useful contribution to heuristic qubit mapping: the algorithmic components are described clearly enough to reimplement, the complexity bound O(N_i * |E|^l * wnd * |G|) is stated explicitly, and the arithmetic in Table 2 is internally consistent with the claimed 20.62% improvement over the TWP column. The HAIL-imp idea of partially extending low-depth sequences is a sensible engineering trade-off. However, the significance is currently conditional: the baseline results are not documented as being produced under the same controlled conditions, and the hyperparameters appear to be tuned on the same benchmarks used for evaluation. These issues affect the central claim of superiority and should be resolved before the results can be accepted as stated.

major comments (3)
  1. [§4, Comparative Algorithms; Table 2] The paper never states how the baseline columns g1, g2, and g3 for SABRE, ILS, and TWP were produced. Were these numbers generated by the authors' own implementations or by rerunning the original code, on the same IBM Q20 coupling graph, with the same benchmark circuit versions, the same gate-counting convention, and the same random seeds or iteration counts? Since SABRE uses random initial mappings, ILS uses randomized perturbations and iteration budgets, and TWP uses look-ahead parameters, externally copied numbers would not constitute a controlled comparison. The aggregated improvements of 74.35%, 26.75%, and 20.62% computed through Eq. (8) are meaningful only under a fully specified, identical protocol. Please provide the baseline generation protocol, the number of runs, and the variance or best/median statistics, and ideally release the code and data so that the comparison can be reproduced.
  2. [Table 1 vs. Table 2] Table 1 reports B23 as having 63,333 total gates and 1,202.609 average CNOTs per circuit, but summing the Gate Num column of Table 2 over the same 23 rows gives 117,289, and 1,202.609 × 23 is approximately 27,660, not 63,333. This internal inconsistency prevents the reader from verifying that the same circuit set was used throughout the comparison and undermines the auditability of the experimental data. Please correct the table or clearly define which quantity is being reported in each column.
  3. [§4, Algorithm Configuration; §4.2] The hyperparameters N_i = 5, the look-ahead depth of 3, the wnd rule (⌊1.5√rg⌋ when rg > 4000 and 30 otherwise), and K = 50 appear to be selected on the same benchmarks used for evaluation, with no sensitivity analysis or validation split. Since the empirical advantage is the paper's main claim, it is important to show how the results vary with these parameters, in particular the search depth, K, and wnd. The paper itself notes negative or unstable cases (qft_10 versus ILS, z4_268 versus TWP), so a systematic parameter-sensitivity study would also clarify whether the reported gains are robust rather than artifacts of a favorable parameter choice.
minor comments (6)
  1. [Abstract; §1] The phrase "optimal initial qubit mapping" overstates the guarantee of the described greedy, subgraph-isomorphism-based heuristic; please replace it with a more qualified term such as "effective" or "good" unless an optimality proof is intended.
  2. [§3.2.1, Eq. (6)] Equation (6) defines Gval in terms of "Executable gate count under τ'", but the paper does not specify whether this counts only front-layer two-qubit gates, gates within the look-ahead window, or some other executable set; please define the executable gate set precisely.
  3. [§3.1.2, Eq. (3)] Equation (3) uses E_w(u,v) as the edge weight between logic qubits corresponding to physical nodes u and v, but at the moment an unmapped qubit is being placed, the logical correspondence of v is not yet fixed; please clarify which edge weight is meant.
  4. [§4.1; Figures 7 and 8] Section 4.1 contains a broken cross-reference ("Table ??"), and Figures 7 and 8 do not report the numeric values plotted or the exact normalization denominators; please fix these presentation issues.
  5. [Throughout; §2.1 and Table 2] The text alternately refers to "SWAP gates" and "CNOT gates" (for example, the abstract versus Table 2), and Section 2.1 notes that a SWAP decomposes into three CNOTs; please state explicitly whether inserted SWAPs are counted as SWAP gates or as their CNOT decomposition throughout the experiments.
  6. [Table 1; §4] The benchmark sets B_ran and B_real are listed in Table 1 without explanation of their provenance; please add references or a description of how these circuit sets were generated.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: HAIL is an empirical heuristic whose claims are grounded in direct benchmark comparisons, not in a self-referential derivation.

full rationale

The paper proposes a heuristic qubit-mapping algorithm (HAIL) and evaluates it against SABRE, ILS, and TWP on standard benchmarks. There is no equation-level derivation in which an output is defined in terms of the quantity it is supposed to predict; the layer-weight formula (Eq. 1), edge-weight formula (Eq. 2), completion scoring function (Eq. 3), SWAP-sequence heuristic (Eq. 6), and post-processing function (Eq. 7) are all constructive heuristics whose parameters are empirical choices. The headline improvement of 20.62% is computed by Eq. (8), Δ_i = 1 − g0/g_i, directly from the paper's reported table columns, so it is a measurement of the algorithm's own output rather than a fitted quantity masquerading as a prediction. No load-bearing claim is justified by a self-citation: the cited prior work is external to the authors, and the paper does not invoke a uniqueness theorem or import an ansatz from the authors' previous publications. The comparatively weak experimental provenance of the baseline numbers and the apparent inconsistency between Table 1's B23 gate total and the sum of Table 2's gate counts are legitimate correctness and reproducibility concerns, but they are not circularity: even if the baselines were miscomputed, the HAIL results would not reduce to the inputs of the algorithm by construction. The central claim is therefore self-contained as an empirical comparison, and no circular step meeting the quoted-evidence standard is present.

Assumptions & free parameters 6 free parameters · 7 assumptions · 0 invented entities

The central claim rests on multiple heuristic design choices rather than a formal derivation. The algorithm's performance is empirical; no new physical entities are introduced. The main free parameters are hyperparameters tuned on the same benchmark set, and the key axioms are heuristic correlations between weights, distances, and SWAP counts that are not proven.

free parameters (6)
  • N_i = 5
    Number of iterations in the forward-backward refinement framework; fixed in experiments without sensitivity analysis.
  • Look-ahead layers = 3
    Number of remaining circuit layers considered when constructing the SWAP candidate set Edges_tp; chosen by hand.
  • wnd (look-ahead window size) = 30, or floor(1.5*sqrt(rg)) when remaining gates rg > 4000
    Window size for the post-processing score d_w; set dynamically based on remaining gate count, tuned on the same benchmark.
  • K = 50
    Number of top depth-2 sequences extended to depth-3 in HAIL-imp; chosen without sensitivity analysis.
  • Search depth l = 2 or 3
    Maximum length of SWAP sequences in the routing search; HAIL-3 uses depth 3, HAIL-2 uses depth 2, chosen as a runtime/quality trade-off.
  • Layer-weight formula = W_g(g_i) = Depth(LC) - Layer(g_i) + 1
    Ad hoc weight assigning higher priority to earlier gates; a design choice that drives the initial mapping and tie-breaking, not fitted to data.
assumptions (7)
  • standard math The qubit mapping problem is NP-complete, justifying heuristic approaches.
    Invoked in the Introduction with citations [9,10]; standard complexity result.
  • domain assumption Single-qubit gates do not affect the qubit mapping outcome and can be ignored.
    Section 2.1 states that gate permutation or cancellation is not performed, so only two-qubit gates matter for SWAP insertion; reasonable for gate-count minimization.
  • standard math The circuit can be partitioned into layers where gates in the same layer act on disjoint qubits.
    Used in Section 3.1.1 to define layer numbers and gate weights; a standard circuit representation.
  • standard math VF2 subgraph isomorphism is correct and complete.
    Used in Algorithm 1 for subgraph matching; VF2 is a well-known exact algorithm cited as [47].
  • ad hoc to paper Assigning higher weights to earlier gates and minimizing weighted distances reduces the total number of SWAPs.
    Core heuristic premise behind Eqs. (1), (3), and (7); plausible but unproven, and central to the algorithm's design.
  • ad hoc to paper Placing unmapped qubits only on nodes adjacent to already placed qubits is sufficient to find a good completion.
    Assumed in Algorithm 2's free_place set; could miss better global placements, but reduces search space.
  • ad hoc to paper The metric 'average executable CNOT count per SWAP' correlates with minimizing the total number of SWAPs.
    Used as the primary routing score G_val in Eq. (6); a heuristic proxy rather than an exact objective.

how reviews work

0 comments
Cite this review

Pith. "Pith review of An Efficient Iterative Algorithm for Qubit Mapping via Layer-Weight Assignment and Search Space Reduction." pith.science (2026). https://pith.science/paper/UBTMOLZ3

@misc{pith2026250207536,
  author       = {Pith},
  title        = {Pith review of: An Efficient Iterative Algorithm for Qubit Mapping via Layer-Weight Assignment and Search Space Reduction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UBTMOLZ3}},
  note         = {Machine review of arXiv:2502.07536}
}
abstract

Current quantum devices support interactions only between physically adjacent qubits, preventing quantum circuits from being directly executed on these devices. Therefore, SWAP gates are required to remap logical qubits to physical qubits, which in turn increases both quantum resource consumption and error rates. To minimize the insertion of additional SWAP gates, we propose HAIL, an efficient iterative qubit mapping algorithm. Leveraging the inherent parallelism in quantum circuits, a new layer-weight assignment method is integrated with subgraph isomorphism to derive an optimal initial qubit mapping. Moreover, we present a two-stage SWAP sequence search algorithm that effectively identifies the most efficient SWAP sequence by distilling feasible SWAP sequences at different stages. The whole qubit mapping algorithm is then refined through a few iterative bidirectional traversals, further reducing the number of SWAP gates required. Experimental results on the IBM Q20 architecture and various benchmarks show that HAIL-3 reduces the number of additional gates inserted in the $\mathcal{B}_{23}$ by 20.62\% compared to state-of-the-art algorithms. Moreover, we propose a partially extended SWAP sequence strategy combined with HAIL to reduce its time complexity, with experiments on the sparsely connected Google Sycamore architecture demonstrating reductions in both algorithm runtime and additional SWAP gates.

Figures

Figures reproduced from arXiv: 2502.07536 by the authors.

Figure 1
Figure 1. Some common basic gates. (a) H gate, (b) Ry( [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. (a) An initial circuit and one of its corresponding [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. The quantum hardware architectures. (a) IBM QX2, [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Paradigm of initial mapping generation. In Step 1, each gate in the quantum circuit is prioritized based on layer-weight [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: Diagram of SWAP sequence insertion. Four CNOT [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: A simplified iterative optimization framework for the whole qubit mapping problem. [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 8
Figure 8. Figure 8: As we can see both the HAIL-imp and HAIL-2 algorithms [PITH_FULL_IMAGE:figures/full_fig_p010_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 45 canonical work pages

  1. [1]

    Simulating physics with computers,

    R. P. Feynman et al., “Simulating physics with computers, ” Int. j. Theor. phys , vol. 21, no. 6/7, 2018

    work page 2018

  2. [2]

    Algorithms for quantum computation: discrete logarithms and fac- toring,

    P. W. Shor, “Algorithms for quantum computation: discrete logarithms and fac- toring, ” inProceedings 35th annual symposium on foundations of computer science . Ieee, 1994, pp. 124–134

    work page 1994

  3. [3]

    A fast quantum mechanical algorithm for database search,

    L. K. Grover, “A fast quantum mechanical algorithm for database search, ” in Proceedings of the twenty-eighth annual ACM symposium on Theory of computing , 1996, pp. 212–219

    work page 1996

  4. [4]

    Quantum algorithm for linear systems of equations,

    A. W. Harrow, A. Hassidim, and S. Lloyd, “Quantum algorithm for linear systems of equations, ”Physical review letters, vol. 103, no. 15, p. 150502, 2009

    work page 2009

  5. [5]

    An efficient quantum proactive incremental learning algorithm,

    L. Li, J. Li, Y. Song, S. Qin, Q. Wen, and F. Gao, “An efficient quantum proactive incremental learning algorithm, ”Science China Physics, Mechanics & Astronomy , vol. 68, no. 1, pp. 1–9, 2025

    work page 2025

  6. [6]

    A quantum federated learning framework for classical clients,

    Y. Song, Y. Wu, S. Wu, D. Li, Q. Wen, S. Qin, and F. Gao, “A quantum federated learning framework for classical clients, ” Science China Physics, Mechanics & Astronomy, vol. 67, no. 5, p. 250311, 2024

    work page 2024

  7. [7]

    Ibm releases first-ever 1,000-qubit quantum chip,

    D. Castelvecchi, “Ibm releases first-ever 1,000-qubit quantum chip, ”Nature, vol. 624, no. 7991, pp. 238–238, 2023

    work page 2023

  8. [8]

    Asymptotically optimal cir- cuit depth for quantum state preparation and general unitary synthesis,

    X. Sun, G. Tian, S. Yang, P. Yuan, and S. Zhang, “Asymptotically optimal cir- cuit depth for quantum state preparation and general unitary synthesis, ”IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems , vol. 42, no. 10, pp. 3301–3314, 2023

    work page 2023

Show all 47 references
  1. [9]

    On the complexity of quantum circuit compilation,

    A. Botea, A. Kishimoto, and R. Marinescu, “On the complexity of quantum circuit compilation, ” inProceedings of the International Symposium on Combinatorial Search, vol. 9, no. 1, 2018, pp. 138–142

    2018

  2. [10]

    Qubit alloca- tion,

    M. Y. Siraichi, V. F. d. Santos, C. Collange, and F. M. Q. Pereira, “Qubit alloca- tion, ” inProceedings of the 2018 International Symposium on Code Generation and Optimization, 2018, pp. 113–125

    2018

  3. [11]

    Quantum circuit synthesis and compilation opti- mization: Overview and prospects,

    Y. Ge, W. Wenjie, C. Yuheng, P. Kaisen, L. Xudong, Z. Zixiang, W. Yuhan, W. Ruocheng, and Y. Junchi, “Quantum circuit synthesis and compilation opti- mization: Overview and prospects, ” 2024. [Online]. Available: https://arxiv.org/ abs/2407.00736

    2024

  4. [12]

    Qubit placement to minimize communi- cation overhead in 2d quantum architectures,

    A. Shafaei, M. Saeedi, and M. Pedram, “Qubit placement to minimize communi- cation overhead in 2d quantum architectures, ” in2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC). IEEE, 2014, pp. 495–500

    2014

  5. [13]

    Optimal qubit assignment and routing via integer programming,

    G. Nannicini, L. S. Bishop, O. Günlük, and P. Jurcevic, “Optimal qubit assignment and routing via integer programming, ”ACM Transactions on Quantum Computing, vol. 4, no. 1, pp. 1–31, 2022

    2022

  6. [14]

    Depth-optimal quantum circuit place- ment for arbitrary topologies,

    D. Bhattacharjee and A. Chattopadhyay, “Depth-optimal quantum circuit place- ment for arbitrary topologies, ”arXiv preprint arXiv:1703.08540, 2017

    2017 arXiv

  7. [15]

    Muqut: Multi-constraint quantum circuit mapping on nisq computers,

    D. Bhattacharjee, A. A. Saki, M. Alam, A. Chattopadhyay, and S. Ghosh, “Muqut: Multi-constraint quantum circuit mapping on nisq computers, ” in2019 IEEE/ACM international conference on computer-aided design (ICCAD) . IEEE, 2019, pp. 1–7

    2019

  8. [16]

    Optimal qubit mapping with simultaneous gate absorption,

    B. Tan and J. Cong, “Optimal qubit mapping with simultaneous gate absorption, ” in 2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD) . IEEE, 2021, pp. 1–8

    2021

  9. [17]

    Noise- adaptive compiler mappings for noisy intermediate-scale quantum computers,

    P. Murali, J. M. Baker, A. Javadi-Abhari, F. T. Chong, and M. Martonosi, “Noise- adaptive compiler mappings for noisy intermediate-scale quantum computers, ” in Proceedings of the twenty-fourth international conference on architectural support for programming languages and ope...

    2019

  10. [18]

    Determining the minimal number of swap gates for multi-dimensional nearest neighbor quantum circuits,

    A. Lye, R. Wille, and R. Drechsler, “Determining the minimal number of swap gates for multi-dimensional nearest neighbor quantum circuits, ” inThe 20th Asia and South Pacific Design Automation Conference . IEEE, 2015, pp. 178–183

    2015

  11. [19]

    Mapping quantum circuits to ibm qx architectures using the minimal number of swap and h operations,

    R. Wille, L. Burgholzer, and A. Zulehner, “Mapping quantum circuits to ibm qx architectures using the minimal number of swap and h operations, ” inProceedings of the 56th Annual Design Automation Conference 2019 , 2019, pp. 1–6

    2019

  12. [20]

    Qubit mapping and routing via maxsat,

    A. Molavi, A. Xu, M. Diges, L. Pick, S. Tannu, and A. Albarghouthi, “Qubit mapping and routing via maxsat, ” in2022 55th IEEE/ACM International Symposium on Microarchitecture (MICRO). IEEE, 2022, pp. 1078–1091

    2022

  13. [21]

    Scalable optimal layout synthesis for nisq quantum processors,

    W.-H. Lin, J. Kimko, B. Tan, N. Bjørner, and J. Cong, “Scalable optimal layout synthesis for nisq quantum processors, ” in2023 60th ACM/IEEE Design Automation Conference (DAC). IEEE, 2023, pp. 1–6

    2023

  14. [22]

    A monte carlo tree search framework for quantum circuit transformation,

    X. Zhou, Y. Feng, and S. Li, “A monte carlo tree search framework for quantum circuit transformation, ” inProceedings of the 39th International Conference on Computer-Aided Design, 2020, pp. 1–7

    2020

  15. [23]

    Qubit routing using graph neural network aided monte carlo tree search,

    A. Sinha, U. Azad, and H. Singh, “Qubit routing using graph neural network aided monte carlo tree search, ” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 36, no. 9, 2022, pp. 9935–9943

    2022

  16. [24]

    Using reinforcement learning to perform qubit routing in quantum compilers,

    M. G. Pozzi, S. J. Herbert, A. Sengupta, and R. D. Mullins, “Using reinforcement learning to perform qubit routing in quantum compilers, ”ACM Transactions on Quantum Computing, vol. 3, no. 2, pp. 1–25, 2022

    2022

  17. [25]

    Deep reinforcement learning for mapping quantum circuits to 2d nearest-neighbor architectures,

    Y. Li, W. Liu, and M. Li, “Deep reinforcement learning for mapping quantum circuits to 2d nearest-neighbor architectures, ”Advanced Quantum Technologies, vol. 7, no. 2, p. 2300289, 2024

    2024

  18. [26]

    Optimizing quantum circuit placement via machine learning,

    H. Fan, C. Guo, and W. Luk, “Optimizing quantum circuit placement via machine learning, ” inProceedings of the 59th ACM/IEEE Design Automation Conference , 2022, pp. 19–24

    2022

  19. [27]

    Supervised learning enhanced quantum circuit trans- formation,

    X. Zhou, Y. Feng, and S. Li, “Supervised learning enhanced quantum circuit trans- formation, ”IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 42, no. 2, pp. 437–447, 2022

    2022

  20. [28]

    Machine learning optimization of quantum circuit layouts,

    A. Paler, L. Sasu, A.-C. Florea, and R. Andonie, “Machine learning optimization of quantum circuit layouts, ”ACM Transactions on Quantum Computing , vol. 4, no. 2, pp. 1–25, 2023

    2023

  21. [29]

    Tackling the qubit mapping problem for nisq-era quantum devices,

    G. Li, Y. Ding, and Y. Xie, “Tackling the qubit mapping problem for nisq-era quantum devices, ” inProceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems , 2019, pp. 1001–1014

    2019

  22. [30]

    Qubit mapping based on subgraph isomorphism and filtered depth-limited search,

    S. Li, X. Zhou, and Y. Feng, “Qubit mapping based on subgraph isomorphism and filtered depth-limited search, ”IEEE Transactions on Computers , vol. 70, no. 11, pp. 1777–1788, 2020

    2020

  23. [31]

    A method based on timing weight priority and distance optimization for quantum circuit transformation,

    Y. Qian, Z. Guan, S. Zheng, and S. Feng, “A method based on timing weight priority and distance optimization for quantum circuit transformation, ”Entropy, vol. 25, no. 3, p. 465, 2023

    2023

  24. [32]

    An efficient methodology for mapping quantum circuits to the ibm qx architectures,

    A. Zulehner, A. Paler, and R. Wille, “An efficient methodology for mapping quantum circuits to the ibm qx architectures, ”IEEE Transactions on Computer- Aided Design of Integrated Circuits and Systems , vol. 38, no. 7, pp. 1226–1236, Conference’17, July 2017, Washington, DC, U...

    2017

  25. [33]

    An iterated local search methodology for the qubit mapping problem,

    P. Zhu, S. Feng, and Z. Guan, “An iterated local search methodology for the qubit mapping problem, ”IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 41, no. 8, pp. 2587–2597, 2021

    2021

  26. [34]

    A dynamic look-ahead heuristic for the qubit mapping problem of nisq computers,

    P. Zhu, Z. Guan, and X. Cheng, “A dynamic look-ahead heuristic for the qubit mapping problem of nisq computers, ”IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems , vol. 39, no. 12, pp. 4721–4735, 2020

    2020

  27. [35]

    Changing the gate order for optimal lnn conversion,

    A. Matsuo and S. Yamashita, “Changing the gate order for optimal lnn conversion, ” in Reversible Computation: Third International Workshop, RC 2011, Gent, Belgium, July 4-5, 2011. Revised Papers 3 . Springer, 2012, pp. 89–101

    2011

  28. [36]

    Qure: Qubit re-allocation in noisy intermediate-scale quantum computers,

    A. Ash-Saki, M. Alam, and S. Ghosh, “Qure: Qubit re-allocation in noisy intermediate-scale quantum computers, ” inProceedings of the 56th Annual Design Automation Conference 2019, 2019, pp. 1–6

    2019

  29. [37]

    A hardware-aware heuristic for the qubit mapping problem in the nisq era,

    S. Niu, A. Suau, G. Staffelbach, and A. Todri-Sanial, “A hardware-aware heuristic for the qubit mapping problem in the nisq era, ”IEEE Transactions on Quantum Engineering, vol. 1, pp. 1–14, 2020

    2020

  30. [38]

    2qan: A quantum compiler for 2-local qubit hamiltonian simulation algorithms,

    L. Lao and D. E. Browne, “2qan: A quantum compiler for 2-local qubit hamiltonian simulation algorithms, ” inProceedings of the 49th Annual International Symposium on Computer Architecture, 2022, pp. 351–365

    2022

  31. [39]

    Timing-aware qubit mapping and gate scheduling adapted to neutral atom quantum computing,

    Y. Li, Y. Zhang, M. Chen, X. Li, and P. Xu, “Timing-aware qubit mapping and gate scheduling adapted to neutral atom quantum computing, ”IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems , 2023

    2023

  32. [40]

    Quantum circuit transformation based on tabu search,

    H. Jiang, Y. Deng, and M. Xu, “Quantum circuit transformation based on tabu search, ”arXiv preprint arXiv:2104.05214, 2021

    2021 arXiv

  33. [41]

    Improved cost-metric for nearest neighbor mapping of quantum circuits to 2-dimensional hexagonal architecture,

    K. Datta, A. Kole, I. Sengupta, and R. Drechsler, “Improved cost-metric for nearest neighbor mapping of quantum circuits to 2-dimensional hexagonal architecture, ” in International Conference on Reversible Computation . Springer, 2023, pp. 218– 231

    2023

  34. [42]

    Mapping nearest neighbor compliant quantum circuits onto a 2-d hexagonal architecture,

    K.-Y. Chang and C.-Y. Lee, “Mapping nearest neighbor compliant quantum circuits onto a 2-d hexagonal architecture, ”IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems , vol. 41, no. 10, pp. 3373–3386, 2021

    2021

  35. [43]

    Mapping quantum algorithms to multi-core quantum computing architectures,

    A. Ovide, S. Rodrigo, M. Bandic, H. Van Someren, S. Feld, S. Abadal, E. Alarcon, and C. G. Almudever, “Mapping quantum algorithms to multi-core quantum computing architectures, ”arXiv preprint arXiv:2303.16125, 2023

    2023 arXiv

  36. [44]

    Tackling the qubit mapping problem with permutation-aware synthesis,

    J. Liu, E. Younis, M. Weiden, P. Hovland, J. Kubiatowicz, and C. Iancu, “Tackling the qubit mapping problem with permutation-aware synthesis, ” in 2023 IEEE International Conference on Quantum Computing and Engineering (QCE) , vol. 1. IEEE, 2023, pp. 745–756

    2023

  37. [45]

    Efficient qubit routing using a dynamically-extract- and-route framework,

    C.-Y. Huang and W.-K. Mak, “Efficient qubit routing using a dynamically-extract- and-route framework, ”IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2024

    2024

  38. [46]

    Optimality study of existing quantum computing layout synthesis tools,

    B. Tan and J. Cong, “Optimality study of existing quantum computing layout synthesis tools, ”IEEE Transactions on Computers , vol. 70, no. 9, pp. 1363–1373, 2020

    2020

  39. [47]

    A (sub)graph isomorphism algorithm for matching large graphs,

    L. Cordella, P. Foggia, C. Sansone, and M. Vento, “A (sub)graph isomorphism algorithm for matching large graphs, ”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 26, no. 10, pp. 1367–1372, 2004

    2004

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.