pith. sign in

arxiv: 2509.08351 · v3 · submitted 2025-09-10 · 🪐 quant-ph

Generative quantum eigensolver with constrained circuit-cutting overhead

Junya Nakamura , Shinichiro Sanji This is my paper

Pith reviewed 2026-05-18 18:27 UTC · model grok-4.3

classification 🪐 quant-ph
keywords generative quantum eigensolverquantum circuit cuttingground state approximationtransformer decodermolecular ground statesvariational quantum algorithmsquantum machine learning
0
0 comments X

The pith

A generative model for quantum circuits can be trained to keep quantum circuit cutting overhead below a chosen limit while still approximating ground states.

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

The work extends the generative quantum eigensolver so the model produces only circuits whose overhead under quantum circuit cutting stays within a fixed upper bound. This change keeps the original goal of finding low-energy approximations for molecules while allowing the circuits to run on hardware with too few qubits for the full circuit. A new loss function and a hybrid online and offline training procedure help the model balance energy quality against the overhead limit. Experiments on the BeH2 molecule in simulation show that the constrained version still reaches useful energies. The result pairs two techniques that had previously been used separately.

Core claim

By training a transformer decoder to generate quantum circuits whose quantum circuit cutting overhead is upper-bounded, the generative quantum eigensolver continues to produce circuits that approximate molecular ground states while enabling execution through circuit cutting on smaller devices.

What carries the argument

Transformer decoder trained with a penalty term on circuit-cutting overhead plus a hybrid online/offline training schedule to generate circuits that satisfy both the energy objective and the overhead bound.

If this is right

  • Ground-state searches for molecules become possible on quantum processors whose qubit count is smaller than the circuit size.
  • Quantum circuit cutting moves from a theoretical tool to one that can be paired directly with generative circuit design.
  • The same overhead-constrained training can be applied to other variational quantum algorithms that produce circuits.
  • Convergence speed and final accuracy improve when the new loss and hybrid training are used.

Where Pith is reading between the lines

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

  • The constraint technique could be tested on larger molecules to check whether the overhead bound scales without eliminating all low-energy circuits.
  • Similar bounds might be added for other resource costs such as gate count or depth in future generative models.
  • Running the method on actual hardware with realistic noise would reveal whether the reduced overhead also reduces error accumulation.

Load-bearing premise

The model can still locate circuits that give good ground-state approximations even after the overhead bound removes some otherwise promising candidates.

What would settle it

Compare the lowest energies reached by the constrained model and the original unconstrained model on BeH2; if the constrained energies are markedly higher while the overhead bound is enforced, the claim that useful circuits remain available would be false.

Figures

Figures reproduced from arXiv: 2509.08351 by Junya Nakamura, Shinichiro Sanji.

Figure 1
Figure 1. Figure 1: Procedure to compute the log probability for [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Mean absolute values of the gradient ∇θLP−DPO are plotted as a function of z. The results of DPO (α = 0), P-DPO (α = 0.25) and P-DPO (α = 0.5) are shown in the panel (a), (b) and (c), respectively. It is observed that the gradient approaches zero as z increases in DPO, whereas the minimum gradient value departs from zero as α increases. is shown by a solid line, and the region spanning the maximum and mini… view at source ↗
Figure 3
Figure 3. Figure 3: The minimum energy values in unit of Hartree achieved up to each iteration step [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The dependence on the hyperparameter [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: The hybrid approaches with two different sets of the hyperparameters, namely [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Coefficient of variation (CV) during the training period of 3000 steps for the mean [PITH_FULL_IMAGE:figures/full_fig_p013_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: An illustration of a quantum circuit generated by GQE under constraints on the [PITH_FULL_IMAGE:figures/full_fig_p013_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: The plotting rules are the same as those in Figure 3. The result obtained with [PITH_FULL_IMAGE:figures/full_fig_p014_8.png] view at source ↗
read the original abstract

Generative quantum eigensolver (GQE) is a hybrid quantum-classical algorithm that iteratively trains a classical generative machine learning model such that the model can generate quantum circuits with desired properties such as approximating molecular ground states. It offers as many potential applications and as much flexibility as variational quantum eigensolvers, while avoiding the problem of barren plateaus. Quantum circuit cutting (QCC) is a technique to perform quantum computations that require more qubits than available on single quantum devices. It comes with considerable sampling overhead depending on the structure of the circuit to be cut and how the circuit is cut. To make QCC practical, therefore, the circuits to be cut must be designed such that their execution is meaningful and QCC overhead is kept small. In this work, we extend GQE such that the generative model only produces circuits whose overhead by QCC is upper-bounded, while retaining the original purpose of GQE. Consequently, our proposal not only enhances the applicability of GQE through the use of QCC, but also provides a practical application for QCC. Using a transformer decoder implementation of GQE, we evaluate our method through simulated ground state search experiments on the BeH_2 molecule. A new loss function and a hybrid online/offline training strategy are also introduced and it is observed that these tools improve convergence and final energy values.

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 extends the Generative Quantum Eigensolver (GQE) by constraining a transformer-decoder generative model to produce only circuits whose quantum circuit cutting (QCC) overhead is upper-bounded. It introduces a new loss term and a hybrid online/offline training strategy to enforce this bound while preserving the goal of generating circuits that approximate molecular ground states. The approach is evaluated via simulated ground-state searches on the BeH₂ molecule, with the abstract claiming improved convergence and final energies relative to prior GQE variants.

Significance. If the empirical results hold under quantitative scrutiny, the work would provide a concrete mechanism for making GQE compatible with limited-qubit hardware via QCC while keeping overhead controlled, thereby increasing the practical reach of generative quantum algorithms. The hybrid training strategy and constrained loss could also serve as a template for other generative models in quantum circuit design.

major comments (2)
  1. [Abstract and §4] Abstract and §4 (BeH₂ experiments): the claim of improved convergence and final energies is stated without quantitative baseline comparisons to unconstrained GQE, without reported error bars or success rates across runs, and without statistics on the fraction of generated circuits rejected by the overhead bound or how the bound is enforced (hard mask versus soft penalty). These omissions leave the central claim that the constraint retains competitive ground-state approximations only partially supported.
  2. [§3.2] §3.2 (training strategy): the hybrid online/offline procedure and the precise form of the new loss term that incorporates the overhead upper bound are described at a high level, but the manuscript does not specify how the bound is sampled during generation or whether the constraint is applied as a hard filter or a differentiable penalty; this detail is load-bearing for assessing whether the method avoids pruning high-performing regions of circuit space.
minor comments (2)
  1. [§2] Notation for the overhead bound and the QCC cost function should be introduced with an explicit equation early in §2 to avoid ambiguity when the loss term is defined later.
  2. [Figures in §4] Figure captions for the BeH₂ energy convergence plots should include the exact overhead bound value used and the number of independent training runs.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed comments. We address each major point below and will revise the manuscript to strengthen the empirical support and methodological clarity.

read point-by-point responses

  1. Referee: [Abstract and §4] Abstract and §4 (BeH₂ experiments): the claim of improved convergence and final energies is stated without quantitative baseline comparisons to unconstrained GQE, without reported error bars or success rates across runs, and without statistics on the fraction of generated circuits rejected by the overhead bound or how the bound is enforced (hard mask versus soft penalty). These omissions leave the central claim that the constraint retains competitive ground-state approximations only partially supported.

    Authors: We agree that the current presentation would benefit from stronger quantitative support. In the revised manuscript we will add direct numerical comparisons to the unconstrained GQE baseline (including energy values and convergence curves), report standard deviations and success rates over multiple independent runs, and include statistics on the fraction of generated circuits that satisfy the overhead bound. We will also clarify that the bound is enforced primarily through a differentiable soft penalty in the loss (with a tunable coefficient) supplemented by a hard rejection step only in the final selection phase of the hybrid strategy; this combination preserves gradient flow to high-performing circuits while respecting the overhead limit. revision: yes

  2. Referee: [§3.2] §3.2 (training strategy): the hybrid online/offline procedure and the precise form of the new loss term that incorporates the overhead upper bound are described at a high level, but the manuscript does not specify how the bound is sampled during generation or whether the constraint is applied as a hard filter or a differentiable penalty; this detail is load-bearing for assessing whether the method avoids pruning high-performing regions of circuit space.

    Authors: We thank the referee for identifying this important clarification. The overhead bound is incorporated as a differentiable soft-penalty term added to the original GQE loss; the penalty is computed on-the-fly for each sampled circuit during the online phase of training. Circuits are generated autoregressively by the transformer decoder, their QCC overhead is evaluated exactly, and the penalty is applied proportionally to the excess overhead. The offline phase then fine-tunes on a curated set of low-overhead circuits. Because the penalty is differentiable, gradients continue to flow toward promising circuit regions even when the bound is temporarily violated. We will expand §3.2 with the explicit loss formula, the sampling procedure, and a short algorithmic outline to make these mechanics fully reproducible. revision: yes

Circularity Check

0 steps flagged

No significant circularity; new constraint, loss, and training strategy are independently introduced

full rationale

The paper extends the existing GQE framework by defining a new overhead bound on QCC, a custom loss function that penalizes violations of this bound, and a hybrid online/offline training procedure for the transformer decoder. These elements are presented as novel additions rather than re-derivations of prior fitted parameters or self-cited uniqueness results. The central claims rest on empirical observations from simulated BeH2 ground-state searches, which are described as verifiable through the reported experiments and not forced by construction from the inputs. No load-bearing step reduces to a self-definition, renamed known result, or self-citation chain that would make the outcome tautological.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The method depends on standard quantum circuit cutting overhead formulas and on the capacity of a transformer-based generative model to learn under an added constraint; no new physical entities are postulated.

free parameters (1)
  • overhead upper bound
    A tunable limit on allowed circuit-cutting overhead that is chosen to trade off between circuit quality and sampling cost.
axioms (1)
  • domain assumption Quantum circuit cutting overhead is determined by the number and type of cuts and can be computed from circuit structure
    The constraint mechanism presupposes that overhead can be evaluated or estimated for each generated circuit during training.

pith-pipeline@v0.9.0 · 5766 in / 1374 out tokens · 45005 ms · 2026-05-18T18:27:26.874423+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.

Reference graph

Works this paper leans on

40 extracted references · 40 canonical work pages · 1 internal anchor

  1. [1]

    The generative quantum eigensolver (GQE) and its application for ground state search

    Kouhei Nakaji, Lasse Bjørn Kristensen, Jorge A Campos-Gonzalez-Angulo, Moham- mad Ghazi Vakili, Haozhe Huang, Mohsen Bagherimehrab, Christoph Gorgulla, FuTe Wong, Alex McCaskey, Jin-Sung Kim, et al. The generative quantum eigensolver (gqe) and its application for ground state search. arXiv preprint arXiv:2401.09253 , 2024

    work page arXiv 2024

  2. [2]

    Variational quantum algorithms

    Marco Cerezo, Andrew Arrasmith, Ryan Babbush, Simon C Benjamin, Suguru Endo, Keisuke Fujii, Jarrod R McClean, Kosuke Mitarai, Xiao Yuan, Lukasz Cincio, et al. Variational quantum algorithms. Nature Reviews Physics, 3(9):625–644, 2021

    work page 2021

  3. [3]

    From transistor to trapped-ion computers for quan- tum chemistry

    M-H Yung, Jorge Casanova, Antonio Mezzacapo, Jarrod Mcclean, Lucas Lamata, Alan Aspuru-Guzik, and Enrique Solano. From transistor to trapped-ion computers for quan- tum chemistry. Scientific reports, 4(1):3589, 2014

    work page 2014

  4. [4]

    A variational eigenvalue solver on a photonic quantum processor

    Alberto Peruzzo, Jarrod McClean, Peter Shadbolt, Man-Hong Yung, Xiao-Qi Zhou, Peter J Love, Al´ an Aspuru-Guzik, and Jeremy L O’brien. A variational eigenvalue solver on a photonic quantum processor. Nature communications, 5(1):4213, 2014

    work page 2014

  5. [5]

    A Quantum Approximate Optimization Algorithm

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

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  6. [6]

    Quantum circuit learning

    Kosuke Mitarai, Makoto Negoro, Masahiro Kitagawa, and Keisuke Fujii. Quantum circuit learning. Physical Review A, 98(3):032309, 2018

    work page 2018

  7. [7]

    Quantum computing in the nisq era and beyond

    John Preskill. Quantum computing in the nisq era and beyond. Quantum, 2:79, 2018

    work page 2018

  8. [8]

    Quantum error mitigation as a universal error reduction technique: Applications from the nisq to the fault-tolerant quantum computing eras

    Yasunari Suzuki, Suguru Endo, Keisuke Fujii, and Yuuki Tokunaga. Quantum error mitigation as a universal error reduction technique: Applications from the nisq to the fault-tolerant quantum computing eras. PRX Quantum, 3(1):010345, 2022

    work page 2022

  9. [9]

    Barren plateaus in quantum neural network training landscapes

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

    work page 2018

  10. [10]

    Noise-induced barren plateaus in variational quantum algorithms

    Samson Wang, Enrico Fontana, Marco Cerezo, Kunal Sharma, Akira Sone, Lukasz Cincio, and Patrick J Coles. Noise-induced barren plateaus in variational quantum algorithms. Nature communications, 12(1):6961, 2021. 16

    work page 2021

  11. [11]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems, 30, 2017

    work page 2017

  12. [12]

    Language models are unsupervised multitask learners

    Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. Language models are unsupervised multitask learners. OpenAI, 2019. Accessed: 2024- 11-15

    work page 2019

  13. [13]

    An image is worth 16x16 words: Trans- formers for image recognition at scale, 2021

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Syl- vain Gelly, Jakob Uszkoreit, and Neil Houlsby. An image is worth 16x16 words: Trans- formers for image recognition at scale, 2021

    work page 2021

  14. [14]

    Point trans- former, 2021

    Hengshuang Zhao, Li Jiang, Jiaya Jia, Philip Torr, and Vladlen Koltun. Point trans- former, 2021

    work page 2021

  15. [15]

    Quantum circuit generation for amplitude encoding using a transformer decoder

    Shunsuke Daimon and Yu-ichiro Matsushita. Quantum circuit generation for amplitude encoding using a transformer decoder. Physical Review Applied, 22(4):L041001, 2024

    work page 2024

  16. [16]

    Minami, K

    Shunya Minami, Kouhei Nakaji, Yohichi Suzuki, Al´ an Aspuru-Guzik, and Tadashi Kad- owaki. Generative quantum combinatorial optimization by means of a novel conditional generative quantum eigensolver. arXiv preprint arXiv:2501.16986 , 2025

    work page arXiv 2025

  17. [17]

    QAOA-GPT: Efficient Generation of Adaptive and Regular Quantum Approximate Optimization Algorithm Circuits,

    Ilya Tyagin, Marwa H Farag, Kyle Sherbert, Karunya Shirali, Yuri Alexeev, and Ilya Safro. Qaoa-gpt: Efficient generation of adaptive and regular quantum approximate optimization algorithm circuits. arXiv preprint arXiv:2504.16350 , 2025

    work page arXiv 2025

  18. [18]

    Optimizing ansatz design in quantum generative adversarial networks using large language models

    Kento Ueda and Atsushi Matsuo. Optimizing ansatz design in quantum generative adversarial networks using large language models. arXiv preprint arXiv:2503.12884 , 2025

    work page arXiv 2025

  19. [19]

    Fine-tuning large language models on quantum optimization problems for circuit generation.arXiv preprint arXiv:2504.11109, 2025

    Linus Jern, Valter Uotila, Cong Yu, and Bo Zhao. Fine-tuning large language models on quantum optimization problems for circuit generation.arXiv preprint arXiv:2504.11109, 2025

    work page arXiv 2025

  20. [20]

    Manning, and Chelsea Finn

    Rafael Rafailov, Archit Sharma, Eric Mitchell, Stefano Ermon, Christopher D. Manning, and Chelsea Finn. Direct preference optimization: Your language model is secretly a reward model, 2024

    work page 2024

  21. [21]

    Contrastive preference optimization: Pushing the boundaries of llm performance in machine translation, 2024

    Haoran Xu, Amr Sharaf, Yunmo Chen, Weiting Tan, Lingfeng Shen, Benjamin Van Durme, Kenton Murray, and Young Jin Kim. Contrastive preference optimization: Pushing the boundaries of llm performance in machine translation, 2024

    work page 2024

  22. [22]

    Harrow, Maris Ozols, and Xiaodi Wu

    Tianyi Peng, Aram W. Harrow, Maris Ozols, and Xiaodi Wu. Simulating large quantum circuits on a small quantum computer. Physical Review Letters, 125(15), oct 2020

    work page 2020

  23. [23]

    Constructing a virtual two-qubit gate by sampling single-qubit operations

    Kosuke Mitarai and Keisuke Fujii. Constructing a virtual two-qubit gate by sampling single-qubit operations. New Journal of Physics , 23(2):023021, feb 2021. 17

    work page 2021

  24. [24]

    Simpo: Simple preference optimization with a reference-free reward, 2024

    Yu Meng, Mengzhou Xia, and Danqi Chen. Simpo: Simple preference optimization with a reference-free reward, 2024

    work page 2024

  25. [25]

    Reinforcement Learning for Robots Using Neural Networks

    Long-Ji Lin. Reinforcement Learning for Robots Using Neural Networks . PhD thesis, School of Computer Science, Carnegie Mellon University, Pittsburg, USA, 1993

    work page 1993

  26. [26]

    Human-level control through deep reinforcement learning

    Volodymyr Mnih, Koray Kavukcuoglu, David Silver, Andrei A Rusu, Joel Veness, Marc G Bellemare, Alex Graves, Martin Riedmiller, Andreas K Fidjeland, Georg Ostrovski, et al. Human-level control through deep reinforcement learning. nature, 518(7540):529–533, 2015

    work page 2015

  27. [27]

    Pytorch: An imperative style, high-performance deep learning library, 2019

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, Alban Desmai- son, Andreas K¨ opf, Edward Yang, Zach DeVito, Martin Raison, Alykhan Tejani, Sasank Chilamkurthy, Benoit Steiner, Lu Fang, Junjie Bai, and Soumith Chintala. Pytorch: An imperative style, high-perf...

    work page 2019

  28. [28]

    Build A Large Language Model (From Scratch)

    Sebastian Raschka. Build A Large Language Model (From Scratch) . Manning, 2024

    work page 2024

  29. [29]

    Pennylane quantum chemistry datasets

    Utkarsh Azad and Stepan Fomichev. Pennylane quantum chemistry datasets. https: //pennylane.ai/datasets/beh2-molecule, 2023

    work page 2023

  30. [30]

    Local, expressive, quantum-number-preserving vqe ans¨ atze for fermionic systems.New Journal of Physics , 23(11):113010, November 2021

    Gian-Luca R Anselmetti, David Wierichs, Christian Gogolin, and Robert M Parrish. Local, expressive, quantum-number-preserving vqe ans¨ atze for fermionic systems.New Journal of Physics , 23(11):113010, November 2021

    work page 2021

  31. [31]

    Sohaib Alam, Guillermo Alonso-Linaje, B

    Ville Bergholm, Josh Izaac, Maria Schuld, Christian Gogolin, Shahnawaz Ahmed, Vishnu Ajith, M. Sohaib Alam, Guillermo Alonso-Linaje, B. AkashNarayanan, Ali Asadi, Juan Miguel Arrazola, Utkarsh Azad, Sam Banning, Carsten Blank, Thomas R Bromley, Benjamin A. Cordier, Jack Ceroni, Alain Delgado, Olivia Di Matteo, Amintor Dusko, Tanya Garg, Diego Guala, Antho...

    work page 2022

  32. [32]

    Hybrid quantum programming with PennyLane Lightning on HPC platforms, 2024

    Ali Asadi, Amintor Dusko, Chae-Yeun Park, Vincent Michaud-Rioux, Isidor Schoch, Shuli Shu, Trevor Vincent, and Lee James O’Riordan. Hybrid quantum programming with PennyLane Lightning on HPC platforms, 2024

    work page 2024

  33. [33]

    Overhead for simulating a non-local channel with local channels by quasiprobability sampling

    Kosuke Mitarai and Keisuke Fujii. Overhead for simulating a non-local channel with local channels by quasiprobability sampling. Quantum, 5:388, 2021. 18

    work page 2021

  34. [34]

    Circuit knitting with classical communication

    Christophe Piveteau and David Sutter. Circuit knitting with classical communication. IEEE Transactions on Information Theory , 70(4):2734–2745, April 2024

    work page 2024

  35. [35]

    Scherer, Axel Plinge, and Christopher Mutschler

    Christian Ufrecht, Maniraman Periyasamy, Sebastian Rietsch, Daniel D. Scherer, Axel Plinge, and Christopher Mutschler. Cutting multi-control quantum gates with zx cal- culus. Quantum, 7:1147, October 2023

    work page 2023

  36. [36]

    Optimal joint cutting of two-qubit rotation gates

    Christian Ufrecht, Laura S Herzog, Daniel D Scherer, Maniraman Periyasamy, Sebastian Rietsch, Axel Plinge, and Christopher Mutschler. Optimal joint cutting of two-qubit rotation gates. Physical Review A, 109(5):052440, 2024

    work page 2024

  37. [37]

    Harrow and Angus Lowe

    Aram W. Harrow and Angus Lowe. Optimal quantum circuit cuts with application to clustered hamiltonian simulation. PRX Quantum, 6(1), January 2025

    work page 2025

  38. [38]

    Cutting circuits with multiple two-qubit unitaries

    Lukas Schmitt, Christophe Piveteau, and David Sutter. Cutting circuits with multiple two-qubit unitaries. Quantum, 9:1634, 2025

    work page 2025

  39. [39]

    Improved sampling bounds and scalable partitioning for quantum circuit cutting beyond bipartitions

    Junya Nakamura, Takahiko Satoh, and Shinichiro Sanji. Improved sampling bounds and scalable partitioning for quantum circuit cutting beyond bipartitions. Phys. Rev. A, 112(4):042422, 2025

    work page 2025

  40. [40]

    Prioritized experience replay, 2016

    Tom Schaul, John Quan, Ioannis Antonoglou, and David Silver. Prioritized experience replay, 2016. 19

    work page 2016