Pith. sign in

REVIEW 3 major objections 5 minor 21 references

Distributed quantum computing can cut batch completion time roughly in half by pairing EPR-aware job ordering with a node-selection step that assigns jobs to the best-connected QPUs, in the simulated network studied here.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-03 02:36 UTC pith:KURWL7VS

load-bearing objection Useful extension of their DQC scheduling framework, but the headline makespan gain from node selection rests on a simulation-level coupling the paper never actually specifies. the 3 major comments →

arxiv 2602.24152 v3 pith:KURWL7VS submitted 2026-02-27 quant-ph cs.DCcs.PF

Advanced Scheduling Strategies for Distributed Quantum Computing Jobs

Gongyu Ni , Davide Ferrari , Lester Ho , Michele Amoretti This is my paper
classification quant-ph cs.DCcs.PF MSC 68M2081P68
keywords distributed quantum computingjob schedulingEPR pairnode selectionmakespanQPU utilizationreinforcement learningPPO
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This paper argues that in a fully connected network of quantum processors with heterogeneous link quality, how you assign arriving DQC jobs to nodes matters almost as much as which jobs you run first. It proposes and simulates several schedulers, and reports that schedulers which order jobs by EPR-pair requirement and then select QPU groups connected by the best links achieve the lowest makespan, roughly half that of FIFO at moderate load. A scheduler that releases nodes asynchronously (ASAP) comes second, and a resource-maximizing heuristic best fills QPUs. The same EPR ordering, without node selection, gives the best per-job latency and fairness. A sympathetic reader would care because DQC networks are expected to have heterogeneous entanglement-generation rates, and this is a concrete, tested comparison of allocation policies.

Core claim

Central claim: link-aware node selection, combined with EPR-aware job ordering, is the main lever for reducing batch makespan in distributed quantum computing. Simulations of a fully connected QPU network with heterogeneous link quality show both EPR (node selection) and PPO (node selection) give the lowest makespan across loads and job-type biases; at λ=4, EPR-NS averages about 7.40e9 ns vs FIFO's 1.47e10 ns. The authors attribute this to assigning jobs to QPU groups whose links have lower state delay (t_state = t_cycle/P_s), speeding entanglement generation. They also report EPR alone gives best latency performance and fairness, Resource-Prioritize gives highest QPU utilization, and ASAP i

What carries the argument

Two mechanisms carry the argument. First, the EPR scheduler reorders jobs by ascending EPR-pair requirement, so jobs that consume fewer non-local entanglement resources run first. Second, the node-selection algorithm picks the K available QPUs whose pairwise link weights have the smallest total, where link quality is captured by expected state delay t_state = t_cycle/P_s; lower t_state means faster entanglement generation. This selection step is what the paper credits for the makespan drop. For the learning variant, a PPO actor groups jobs using a reward that penalizes normalized latency and adds EPR-aware incentives; with node selection the reward also encourages jobs with more EPR pairs to

Load-bearing premise

All reported makespan advantages of node selection rest on the assumption that assigning a job to QPUs connected by lower state-delay links actually shortens that job's execution time in the simulator. The paper never specifies the mapping from link quality to per-job execution time, and Algorithm 3 uses an undefined link weight w(u,v).

What would settle it

Re-run the simulations with per-job execution times sampled from the Qoala dataset independently of the link quality of the assigned node group, keeping all other settings unchanged. If EPR (node selection) still beats plain EPR by roughly the same margin, the link-aware mechanism is not what drives the makespan gain; if the advantage collapses, the mechanism is confirmed.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • A real execution manager for a heterogeneous DQC network should consider link quality when choosing QPUs, not just which jobs run first.
  • Prioritizing low-EPR jobs is the best way to protect per-job latency and fairness, at the cost of leaving some QPUs idle.
  • Resource-maximizing and asynchronous-release schemes keep QPUs busiest but do not win on makespan or fairness.
  • A PPO-based scheduler can approximate hand-designed EPR-aware scheduling, though it still lags slightly under high load.
  • No single scheduler dominates all metrics, so the operator's objective (throughput, utilization, latency, fairness) should determine the choice.

Where Pith is reading between the lines

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

  • If the link-selection effect is real, the gap between EPR (node selection) and plain EPR is a quantitative estimate of what topology-aware allocation is worth; on a homogeneous-link network the two schedulers should coincide.
  • A concrete testable rule follows from the reward design: allocate jobs with the highest EPR demand to the best-quality links, since the reported makespan gains come from that coupling.
  • Because the PPO reward targets latency and EPR usage rather than makespan directly, adding a direct makespan or critical-path term is a natural next experiment, especially where the paper reports PPO lagging EPR-NS at λ=8.
  • The job model is atomic; allowing EPR generation and gate execution to interleave within a job could change which scheduling policy wins.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper addresses execution management in distributed quantum computing (DQC) by proposing and simulating several schedulers for allocating DQC jobs to QPUs in a fully connected, heterogeneous network. The proposed methods include classical heuristics (Resource-Prioritize, EPR, EPR with node selection, ASAP) and reinforcement-learning schedulers based on PPO, benchmarked against FIFO and LIST. Evaluation is performed with the Qoala Simulator over 1000 time slots under varying network load and job selection bias, using metrics of makespan, QPU utilization, non-local gate rate, system execution-latency performance (SELP), and fairness. The headline claim is that EPR and PPO schedulers with node selection achieve the lowest makespan by exploiting heterogeneous link quality, with ASAP second and Resource-Prioritize yielding the highest QPU utilization.

Significance. If the reported results hold, the paper provides a useful comparative study of scheduling strategies for DQC, a relatively underexplored area. The explicit taxonomy of schedulers, the use of a realistic simulator, and the provided data/code repository are strengths. However, the central mechanism responsible for the main result—that node selection based on heterogeneous links reduces job execution time—is not demonstrated at the simulator level. Without that link, the headline makespan advantage is unsupported. The absence of uncertainty quantification and the missing PPO hyperparameters further limit the reliability and reproducibility of the quantitative rankings.

major comments (3)
  1. [§5.3, §5.5, §6.2.1] The central claim that EPR(NS) and PPO(NS) achieve the lowest makespan because node selection leverages heterogeneous network links (Fig. 4, Tables 4–5) requires a simulator-level dependency between the assigned node group's link quality and each job's execution time. The paper never specifies this dependency. §5.5 states that execution times are 'derived from the dataset generated by Qoala Simulator,' but no function T_j(nodes, t_state, P_s) or experimental trace is given. Algorithm 3 takes a graph 'with link weights' and sums w(u,v) over pairs, but w(u,v) is never defined, and it is not explained how the selected group N* maps to a sampled T_j. Algorithm 4 uses T_j independently of node assignment. If T_j is drawn independently of node assignment, the node-selection schedulers cannot benefit from lower t_state, and the reported halving of makespan would be an artifact. Please specify t
  2. [Tables 4–9] All reported metrics are averages over 1000 time slots, but no standard deviations, confidence intervals, or statistical tests are provided. Several conclusions rest on small differences: FIFO vs LIST at λ=4 differ by ~0.6% (1.4675e10 vs 1.4584e10 ns), and EPR vs PPO at λ=4 differ by ~1.8%. The prose asserts consistent rankings (e.g., 'LIST is consistently lower than FIFO', 'PPO is lower than EPR') that are not justified without uncertainty quantification. Add error bars, percentile intervals, or pairwise significance tests to support the comparative claims.
  3. [§5.5] The PPO reward functions (Eqs. 18, 20) and loss function (Eq. 21) contain coefficients ι^(1), ι^(2), γ, β, c_v, c_e, but their values are never reported. Training details such as learning rate, network architecture, number of training steps, and convergence criteria are also absent. Without these, the PPO results are not reproducible from the manuscript. The Zenodo repository may contain code, but the paper should state the values or at least point explicitly to the configuration file.
minor comments (5)
  1. [§4.2–4.3] Sections 4.2 and 4.3 are identical duplicates (Eqs. 4 and 5 present the same QPU utilization definition). Remove the duplicate.
  2. [§6.1.2 and figure captions] The bias parameter is defined as τ in Eq. (22), but Fig. 4–8 captions and Tables 4–9 use 'bias parameter α=0.5'. Align the notation.
  3. [§6.2.1] The text says 'the PPO (node selection) curve is consistently slightly lower than that of EPR (node selection), indicating a higher makespan.' A lower curve indicates a lower makespan; the wording is contradictory.
  4. [Appendix A.5] Typo: 'cricuit' should be 'circuit'.
  5. [§5.2 / §6.2.4] The explanation that EPR schedulers achieve high SELP/fairness relies on the assertion that low-EPR jobs have shorter execution times. This is an empirical assumption about the job dataset that should be verified quantitatively, e.g., by reporting the correlation between EPR demand and execution time.

Circularity Check

0 steps flagged

No significant circularity: the schedulers are simulated and benchmarked against independent FIFO/LIST baselines; design-objective alignment is not a definitional reduction, and the missing node-link coupling is an under-specification, not a circular derivation.

full rationale

The paper's claims are simulation results, not analytic derivations, and no prediction reduces to an input by construction. Resource-Prioritize selects combinations maximizing R_tot and minimizing T_tot (Algorithm 1), and Table 6 reports its high QPU utilization; while this is expected from the design objective, U_QPU (Eq. 4) is an independently computed aggregate over actual p_j, q_j, and makespan, so the result is not forced by the optimization itself. Similarly, the EPR scheduler orders by E_j and the paper explicitly assumes low-EPR jobs are shorter (Section 5.2); its high SELP/fairness (Eqs. 9-11, Tables 8-9) follows from that heuristic, but ELP is defined from measured execution and waiting times, not from E_j. The paper does contain a genuine under-specification that would be load-bearing if wrong: Algorithm 3 takes a graph 'with link weights' and sums w(u,v), but w(u,v) is never defined, and Section 5.5 says job execution times are 'derived from the dataset generated by Qoala Simulator' with no equation linking assigned node-group link quality (t_state, Eq. 2) to T_j. That missing coupling threatens the causal claim in Section 6.2.1 that node selection improves makespan through heterogeneous links, but it is an unsupported empirical/implementation assumption, not a circular reduction: no equation in the paper defines T_j in terms of the selected node group, so the reported advantage is not provably an identity. Self-citations (e.g., [8] for FIFO/LIST and QPU utilization, [5] for compilation) are used as background or as baseline descriptions reproduced in the text; they do not carry the argument and no uniqueness or ansatz is imported from them. Overall, no step exhibits the required quote-and-reduction form of circularity.

Axiom & Free-Parameter Ledger

4 free parameters · 5 axioms · 0 invented entities

The paper is an empirical simulation study; its ledger is dominated by modeling assumptions rather than invented entities or fitted constants. The two free-parameter clusters that matter: (i) unreported PPO reward/loss coefficients that shape the trained policies, and (ii) hand-estimated link cycle times that set the absolute timescale of all makespan numbers. No new physical entities are postulated. The most fragile entries are axioms 1, 2, and 5, which carry the non-local-gate metric, the atomic-job abstraction, and the undefined link weights.

free parameters (4)
  • PPO reward/loss coefficients ι^(1), ι^(2), γ, β, c_v, c_e = not reported in paper
    Eqs. 16-21 define these balancing coefficients; no values, ranges, or tuning procedure are given, yet they determine the trained policy's behavior. Critical for replication.
  • Worst-case link cycle time t_cycle = 1.8e6 ns
    Table 3: estimated from the 230-m experiment of [20] and extrapolated to a 0.1-km link; a hand-picked pessimistic value that sets the overall time scale of every makespan in Tables 4-5.
  • Job-selection bias τ (labeled α in figures/tables) = 0.5 for biased scenarios
    Scenario knob from Eq. (22) used to skew sampling toward high-EPR jobs; the same quantity is renamed α in Table captions and Figures 4-8, which makes the 'bias-0.5' columns ambiguous as written.
  • Node-selection link weight w(u,v) = undefined
    Algorithm 3 consumes link weights to pick node groups, but the paper never specifies how they are computed from P_s, t_cycle, or fidelity (Table 2); the headline makespan result is conditioned on this undefined input.
axioms (5)
  • domain assumption Non-local gates dominate DQC job execution time, so pairwise overlap of job execution times approximates overlap of non-local gate execution.
    Invoked in §4.4 to define the non-local-gate-rate metric U_g; the authors themselves note in §6.2.3 that the T_max normalization counts pairs that cannot actually be parallelized, so only within-row comparisons are valid.
  • domain assumption A DQC job is an atomic task with fixed demand (R_j QPUs, E_j EPR pairs) and execution time sampled from a Qoala dataset; intra-job sub-circuit dependencies are ignored.
    The scheduler abstraction throughout §5 and the simulator pipeline in §3.1, inherited from prior work [8]. Real partitioned circuits have inter-part communication, so the model is an idealization.
  • domain assumption Entanglement success probability model P_s(d) of Eq. (1) with Table 2 parameters from [17] applies to the simulated links.
    Link parameters are taken as inputs from the cited Qoala compilation work [17]; accepted without independent validation in this paper.
  • domain assumption Qoala Simulator [19] execution times faithfully represent DQC job execution.
    All scheduler evaluations and the PPO environment draw T_j from this dataset (§5.5, §6.1); no hardware validation is offered.
  • ad hoc to paper Algorithm 3's link weights w(u,v) exist and encode link quality.
    The node-selection algorithms take an unspecified weight as input; the paper never defines it in terms of Table 2 link parameters, making the headline node-selection result dependent on an undefined quantity.

pith-pipeline@v1.3.0-alltime-deepseek · 17831 in / 23225 out tokens · 230153 ms · 2026-08-03T02:36:03.083922+00:00 · methodology

0 comments
read the original abstract

Distributed quantum computing (DQC) is being actively investigated as a means of scaling the number of qubits across multiple connected quantum devices. This includes quantum circuit compilation and execution management on multiple quantum devices in the network. The latter aspect is very challenging because, while reducing the makespan of job batches remains a relevant objective, novel quantum-specific constraints must be considered, including QPU utilization, non-local gate rate, and the latency associated with queued DQC jobs. In this work, a range of scheduling strategies is proposed, simulated, and evaluated, including heuristics that prioritize resource maximization for QPU utilization, node selection based on heterogeneous network connectivity, asynchronous node release upon job completion, and a scheduling strategy based on reinforcement learning with proximal policy optimization. These approaches are benchmarked against traditional FIFO and LIST schedulers under varying DQC job types and network conditions for the allocation of DQC jobs to devices within a network.

Figures

Figures reproduced from arXiv: 2602.24152 by Davide Ferrari, Gongyu Ni, Lester Ho, Michele Amoretti.

Figure 1
Figure 1. Figure 1: The DQC workflow decomposes a monolithic [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Fully connected QPU network with heteroge [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The PPO framework comprises the environ [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The cumulative probability of makespan in each time slot among different schedulers. [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: The cumulative probability of QPU utilization in each time slot among different schedulers. [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
Figure 8
Figure 8. Figure 8: This is because they prioritize jobs with shorter [PITH_FULL_IMAGE:figures/full_fig_p013_8.png] view at source ↗
Figure 6
Figure 6. Figure 6: The cumulative probability of non-local gate rate in each time slot among different schedulers. [PITH_FULL_IMAGE:figures/full_fig_p014_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: The cumulative probability of SELP in each time slot among different schedulers. [PITH_FULL_IMAGE:figures/full_fig_p015_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: The cumulative probability of fairness in each time slot among different schedulers. [PITH_FULL_IMAGE:figures/full_fig_p016_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: 5-qubit GHZ state preparation. The first qubit [PITH_FULL_IMAGE:figures/full_fig_p018_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: 5-qubit graph state preparation. Each qubit [PITH_FULL_IMAGE:figures/full_fig_p018_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: 5-qubit QAOA circuit with one repeti￾tion. The cost unitary UC (γ) is implemented via two controlled-NOT and one Rc(γ) gates, which is a rota￾tion that depends on the parity of the two qubits.The mixer unitary UB(β) is applied as Rb(2β) rotations on all qubits. |q4⟩ H |q3⟩ R2 H |q2⟩ R3 R2 H |q1⟩ R4 R3 R2 H |q0⟩ R5 R4 R3 R2 H [PITH_FULL_IMAGE:figures/full_fig_p019_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: 5-qubit QFT circuit with controlled-phase [PITH_FULL_IMAGE:figures/full_fig_p019_12.png] view at source ↗

discussion (0)

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

Reference graph

Works this paper leans on

21 extracted references · 4 linked inside Pith

  1. [1]

    Quantum internet: Networking challenges in distributed quantum computing.IEEE Network, 34(1):137–143, 2019

    Angela Sara Cacciapuoti, Marcello Caleffi, Francesco Tafuri, Francesco Saverio Cataliotti, Stefano Gherardini, and Giuseppe Bianchi. Quantum internet: Networking challenges in distributed quantum computing.IEEE Network, 34(1):137–143, 2019

    2019

  2. [2]

    Bell nonlocality.Rev

    Nicolas Brunner, Daniel Cavalcanti, Stefano Piro- nio, Valerio Scarani, and Stephanie Wehner. Bell nonlocality.Rev. Mod. Phys., 86:419–478, Apr 2014

    2014

  3. [3]

    Entangle- ment verification in quantum networks with tam- pered nodes.IEEE Journal on Selected Areas in Communications, 38(3):598–604, 2020

    Michele Amoretti and Stefano Carretta. Entangle- ment verification in quantum networks with tam- pered nodes.IEEE Journal on Selected Areas in Communications, 38(3):598–604, 2020

    2020

  4. [4]

    Distributed quantum computing: a survey.Computer Networks, 254:110672, 2024

    Marcello Caleffi, Michele Amoretti, Davide Ferrari, Jessica Illiano, Antonio Manzalini, and Angela Sara Cacciapuoti. Distributed quantum computing: a survey.Computer Networks, 254:110672, 2024

    2024

  5. [5]

    A modular quantum compilation frame- work for distributed quantum computing.IEEE Transactions on Quantum Engineering, 4:1–13, 2023

    Davide Ferrari, Stefano Carretta, and Michele Amoretti. A modular quantum compilation frame- work for distributed quantum computing.IEEE Transactions on Quantum Engineering, 4:1–13, 2023

    2023

  6. [6]

    Remote gate scheduling in distributed quan- tum computing

    Xu Xu, Yu Liu, Yingling Mao, and Yuanyuan Yang. Remote gate scheduling in distributed quan- tum computing. In2025 IEEE 45th Interna- tional Conference on Distributed Computing Sys- tems (ICDCS), pages 846–856. IEEE, 2025

    2025

  7. [7]

    Optimized compiler for distributed quantum computing.ACM Trans- actions on Quantum Computing, 4(2):1–29, 2023

    Daniele Cuomo, Marcello Caleffi, Kevin Krsulich, Filippo Tramonto, Gabriele Agliardi, Enrico Prati, and Angela Sara Cacciapuoti. Optimized compiler for distributed quantum computing.ACM Trans- actions on Quantum Computing, 4(2):1–29, 2023

    2023

  8. [8]

    Execution management of distributed quantum computing jobs

    Davide Ferrari, Michele Bandini, and Michele Amoretti. Execution management of distributed quantum computing jobs. In2024 IEEE Interna- tional Conference on Quantum Computing and En- gineering (QCE), volume 2, pages 150–154. IEEE, 2024

    2024

  9. [9]

    Distributed quantum computing and net- work control for accelerated vqe.IEEE Transac- tions on Quantum Engineering, 2:1–21, 2021

    Stephen DiAdamo, Marco Ghibaudi, and James Cruise. Distributed quantum computing and net- work control for accelerated vqe.IEEE Transac- tions on Quantum Engineering, 2:1–21, 2021

    2021

  10. [10]

    Quantum algorithms and simulation for parallel and distributed quantum computing

    Rhea Parekh, Andrea Ricciardi, Ahmed Darwish, and Stephen DiAdamo. Quantum algorithms and simulation for parallel and distributed quantum computing. In2021 IEEE/ACM Second Interna- tional Workshop on Quantum Computing Software (QCS), pages 9–19. IEEE, 2021

    2021

  11. [11]

    Scheduling parallel jobs to min- imize the makespan.Journal of Scheduling, 9(5):433–452, 2006

    Berit Johannes. Scheduling parallel jobs to min- imize the makespan.Journal of Scheduling, 9(5):433–452, 2006

    2006

  12. [12]

    Parallel job scheduling with over- head: A benchmark study

    Richard A Dutton, Weizhen Mao, Jie Chen, and William Watson. Parallel job scheduling with over- head: A benchmark study. In2008 international conference on networking, architecture, and stor- age, pages 326–333. IEEE, 2008

    2008

  13. [13]

    Multiproces- sor jobs, preemptive schedules, and one-competitive online algorithms

    Jiˇ r ´ ı Sgall and Gerhard J Woeginger. Multiproces- sor jobs, preemptive schedules, and one-competitive online algorithms. InInternational Workshop on Approximation and Online Algorithms, pages 236–

  14. [14]

    Quincy: fair scheduling for distributed computing clusters

    Michael Isard, Vijayan Prabhakaran, Jon Currey, Udi Wieder, Kunal Talwar, and Andrew Goldberg. Quincy: fair scheduling for distributed computing clusters. InProceedings of the ACM SIGOPS 22nd symposium on Operating systems principles, pages 261–276, 2009

    2009

  15. [15]

    Network operations scheduling for dis- tributed quantum computing

    Nitish K Chandra, Eneet Kaur, and Kaushik P Se- shadreesan. Network operations scheduling for dis- tributed quantum computing. In2024 IEEE 6th In- ternational Conference on Trust, Privacy and Secu- rity in Intelligent Systems, and Applications (TPS- ISA), pages 506–515. IEEE, 2024

    2024

  16. [16]

    Qgroup: Parallel quantum job scheduling using dynamic pro- gramming

    Aaron Orenstein and Vipin Chaudhary. Qgroup: Parallel quantum job scheduling using dynamic pro- gramming. In2024 IEEE International conference on quantum computing and engineering (QCE), vol- ume 1, pages 990–999. IEEE, 2024. 17

    2024

  17. [17]

    Compilation strategies for quantum network programs using qoala.arXiv preprint arXiv:2505.06162, 2025

    Samuel Oslovich, Bart van der Vecht, and Stephanie Wehner. Compilation strategies for quantum network programs using qoala.arXiv preprint arXiv:2505.06162, 2025

    Pith/arXiv arXiv 2025

  18. [18]

    Proximal policy optimization algorithms.arXiv preprint arXiv:1707.06347, 2017

    John Schulman, Filip Wolski, Prafulla Dhari- wal, Alec Radford, and Oleg Klimov. Proximal policy optimization algorithms.arXiv preprint arXiv:1707.06347, 2017

    Pith/arXiv arXiv 2017

  19. [19]

    Qoala: an appli- cation execution environment for quantum internet nodes.arXiv preprint arXiv:2502.17296, 2025

    Bart van der Vecht, Atak Talay Y¨ ucel, Hana Jirovsk´ a, and Stephanie Wehner. Qoala: an appli- cation execution environment for quantum internet nodes.arXiv preprint arXiv:2502.17296, 2025

    Pith/arXiv arXiv 2025

  20. [20]

    Krutyanskiy, M

    V. Krutyanskiy, M. Galli, V. Krcmarsky, S. Baier, D. A. Fioretto, Y. Pu, A. Mazloom, P. Sekatski, M. Canteri, M. Teller, J. Schupp, J. Bate, M. Mer- aner, N. Sangouard, B. P. Lanyon, and T. E. Northup. Entanglement of trapped-ion qubits sep- arated by 230 meters.Phys. Rev. Lett., 130:050803, Feb 2023

    2023

  21. [21]

    Quantum algorithm implementations for beginners.arXiv preprint arXiv:1804.03719, 2018

    Adetokunbo Adedoyin, John Ambrosiano, Petr Anisimov, William Casper, Gopinath Chennupati, Carleton Coffrin, Hristo Djidjev, David Gunter, Satish Karra, Nathan Lemons, et al. Quantum algorithm implementations for beginners.arXiv preprint arXiv:1804.03719, 2018. A Specific DQC job types A.1 GHZ State The Greenberger–Horne–Zeilinger (GHZ) state is an en- tan...

    Pith/arXiv arXiv 2018