pith. sign in

arxiv: 2606.19947 · v1 · pith:KW6VLE6Mnew · submitted 2026-06-18 · 🪐 quant-ph · cs.LG

QMaxCal: Path-Space Regularization for Open Quantum Control via Girsanov's Theorem

Merijn Moody , Zier Mensch , Miranda C. N. Cheng , Peter G. Bolhuis , Max Welling This is my paper

Pith reviewed 2026-06-26 17:33 UTC · model grok-4.3

classification 🪐 quant-ph cs.LG
keywords open quantum systemsquantum controlGirsanov theorempath space regularizationcontinuous monitoringtrajectory distributionsdecoherenceKL divergence
0
0 comments X

The pith

Girsanov's theorem supplies closed-form regularizers that steer open quantum control away from strong decoherence states.

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

The paper tries to establish that measurement records from continuously monitored open quantum systems differ only in a drift term when they share the same decoherence channels, so Girsanov's theorem directly yields a differentiable estimator of the KL divergence between trajectory distributions. Two reference measures turn this estimator into regularizers that penalize the observable consequences of control on the noise channels themselves rather than control amplitude. A sympathetic reader would care because these penalties are distinct from fluence or smoothness penalties and demonstrably raise final-state fidelity while improving robustness to noise-model mismatch on single-qubit, multi-qubit, and hardware-calibrated benchmarks. The regularizers reduce infidelity by up to 50 percent and cut occupation of forbidden states.

Core claim

Instantiating the Girsanov-derived KL estimator with a Wiener reference measure and with a drift-variance reference measure produces two regularizers, KL_W and R_DV, that both drive controlled trajectories toward states where decoherence effects are minimal; when added to gradient-based and reinforcement-learning policies these regularizers outperform the unregularized baselines on final-state fidelity, robustness under noise mismatch, and forbidden-state occupation across single- and multi-qubit test cases and on a multi-qubit chain calibrated to a published IBM Kingston snapshot.

What carries the argument

Girsanov's theorem applied to pairs of measurement records that share decoherence channels but differ only in drift, supplying a closed-form differentiable KL estimator between their trajectory distributions.

If this is right

  • The regularizers reduce infidelity by up to 50 percent relative to unregularized policies.
  • Gains in fidelity increase from 17 percentage points at matched noise to 27 percentage points under 2.5-fold noise mismatch.
  • Both regularizers lower occupation of forbidden states compared with baselines.
  • The same performance ordering holds on a multi-qubit chain whose parameters are taken from a published IBM Kingston snapshot.

Where Pith is reading between the lines

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

  • The same Girsanov construction could be applied to other continuously monitored quantum systems whose trajectory measures admit a drift-only difference under shared noise channels.
  • Because the penalties act on observable trajectory statistics rather than on the control waveform, they may remain effective when the control hardware itself introduces additional distortions.
  • The method opens a route to regularizing policies that must satisfy hard constraints on state occupation by folding those constraints directly into the reference measure.

Load-bearing premise

Measurement records of two evolutions that share the same decoherence channels differ only in their drift term.

What would settle it

A direct numerical check on the IBM Kingston-calibrated chain showing that the reported fidelity gains disappear when the assumed noise model is replaced by the true hardware noise would falsify the robustness claim.

Figures

Figures reproduced from arXiv: 2606.19947 by Max Welling, Merijn Moody, Miranda C. N. Cheng, Peter G. Bolhuis, Zier Mensch.

Figure 1
Figure 1. Figure 1: SSE-trajectory population variance at γT = 2. Mean populations across 128 SSE samples (lines) with ±1σ bands for the baseline (λKLW = λRDV = 0), Wiener KL (λKLW = 5), drift-variance (λRDV = 5), and PPO. Wiener KL contracts the time-integrated population variance from 0.0321 (baseline) to 0.0021, a 15× reduction, consistent with the policy routing the state along kerL where both the SSE drift and noise vani… view at source ↗
Figure 2
Figure 2. Figure 2: Diamond system, γtrain = 2, T = 1. Left, middle: SSE-trajectory populations (mean across 128 samples, ±1σ) for baseline (F = 0.665) and Wiener KL (λKLW = 5, F = 0.834); the regularizer routes population through |d⟩ ∈ kerL rather than the lossy direct coupling. Right: fidelity vs. γtest for policies trained at γ = 2; full statistics in Appendix E.4. Hardware-calibrated case study. We further show that drift… view at source ↗
Figure 3
Figure 3. Figure 3: Site populations across the asymmetry sweep. Columns: ρ ∈ {2, 4, 8}. Rows: drift-variance (λRDV = 0.02, top), baseline (bottom). Each panel shows the single-site excitation population ⟨ψ(t)|(I − σ (i) z )/2|ψ(t)⟩ at sites i ∈ {0, 1, 2, 3} (i.e., the probability that site i holds the excitation); these four observables are the natural readout because the task is excitation transfer. Curves are drawn from a … view at source ↗
Figure 4
Figure 4. Figure 4: Level structure and noise structure of the five benchmarks. Blue double arrows mark coherent control couplings; red squiggly arrows mark decoherence channels (each acts on the density matrix as a Lindblad operator Lk). (a) Amplitude damping: |1⟩ decays to |0⟩, two controls drive arbitrary single-qubit rotations. (b) STIRAP: two stable ground states |g1⟩, |g2⟩ coupled through a lossy excited state |e⟩ that … view at source ↗
Figure 5
Figure 5. Figure 5: Wiener KL integrand Eψ[α(t) 2 ], γT = 2. Trajectory-averaged drift squared on a logarithmic axis. Wiener KL suppresses α toward kerL, reducing the time-integrated value 1 2 R ⟨α 2 ⟩ dt from 0.230 (baseline) to 0.031 — an order-of-magnitude reduction consistent with the trajectory routing seen in [PITH_FULL_IMAGE:figures/full_fig_p018_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: STIRAP, γT = 10. SSE-trajectory populations (mean across 128 samples with ±1σ bands) for baseline (λ=0), Wiener KL (λKLW =1), and PPO. • Solver: EM • Discretised time grid: 256 points per trajectory, snapshot every 500 steps; 256 SSE samples per gradient step. • Regulariser weights: λKLW ∈ {0, 0.5, 5}, λRDV = 0, λflu = 0.001. • Fourier controller: nmodes = 20, init_scale = 0.1. • Seeds: 3 per cell. E.5. 4 … view at source ↗
Figure 7
Figure 7. Figure 7: IBM Kingston q14→q9 evolutions. Per-site excitation populations ⟨ψ(t)|(I − σ (i) z )/2|ψ(t)⟩ (i.e., the probability that site i holds the excitation). Top row: the four policies of [PITH_FULL_IMAGE:figures/full_fig_p022_7.png] view at source ↗
read the original abstract

Reliable quantum control in the presence of decoherence requires policies that combat the effect of environmental noise on the controlled dynamics. Open quantum systems under continuous monitoring generate classical measurement records whose drift depends on the noise experienced by the system; the records of two evolutions sharing the same decoherence channels differ only in this drift, so Girsanov's theorem yields a closed-form, differentiable estimator of the KL divergence between their trajectory distributions. We instantiate this estimator with two physically motivated reference measures, yielding two regularizers that both drive the system toward states where the effects of decoherence are minimal: the Wiener KL (KL_W), which is empirically more effective under certain conditions on the noise model, and the drift-variance regularizer (R_DV), which works for all noise models. Both are qualitatively distinct from existing penalties on control fluence or smoothness: they penalize the observable consequences of control on the decoherence channels rather than the control amplitude itself. The regularizers outperform unregularized gradient-based and reinforcement-learning baselines across a range of open quantum systems -- including single- and multi-qubit benchmarks and a multi-qubit chain calibrated to a published snapshot of the IBM Kingston processor -- along several axes of evaluation: final-state fidelity, robustness to mismatch in the assumed noise model (gains grow from +17 pp at training noise to +27 pp under 2.5x noise mismatch), and occupation of forbidden states. The regularizers reduce infidelity by up to 50%, with ~16% gains on the calibrated IBM Kingston chain.

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 paper claims that Girsanov's theorem supplies a closed-form, differentiable KL estimator between trajectory distributions of two open quantum evolutions that share decoherence channels (differing only in drift), which is instantiated as two regularizers (Wiener KL and drift-variance) for quantum control policies; these are shown empirically to outperform unregularized gradient and RL baselines on fidelity, noise-mismatch robustness, and forbidden-state occupation across single-/multi-qubit benchmarks and a calibrated IBM Kingston multi-qubit chain, with infidelity reductions up to 50%.

Significance. If the central derivation and assumption hold without hidden model-specific corrections, the approach supplies a distinct path-space regularization (penalizing observable decoherence consequences rather than control fluence) with reported robustness gains that grow under noise mismatch; the inclusion of a hardware-calibrated benchmark is a concrete strength for practical relevance in open-system control.

major comments (2)
  1. [Abstract / derivation] Abstract and derivation section: the load-bearing premise that 'the records of two evolutions sharing the same decoherence channels differ only in this drift' is asserted without an explicit Radon-Nikodym derivation or verification for the stochastic master equation trajectories used; if control enters the diffusion coefficients, induces cross-qubit correlations, or produces back-action not reducible to drift shift, the claimed closed-form differentiable KL estimator acquires extra factors and the reported performance numbers (including the IBM Kingston ~16% gain) may not follow directly.
  2. [Empirical results] Empirical evaluation (IBM Kingston chain and multi-qubit benchmarks): the abstract reports gains without error bars, data-exclusion rules, or explicit confirmation that the KL estimator remains free of post-hoc parameter choices or model-specific adjustments; this undermines assessment of whether the +17 pp to +27 pp robustness improvement is reproducible or an artifact of the unverified drift-only assumption.
minor comments (2)
  1. [Method] Notation for the two reference measures (KL_W and R_DV) should be introduced with explicit equations showing how each is obtained from the Girsanov Radon-Nikodym derivative.
  2. [Introduction] The distinction from fluence/smoothness penalties is stated qualitatively; a short table comparing the functional form of each regularizer on a common example would clarify the claimed qualitative difference.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments. The feedback highlights important points on the derivation's rigor and the presentation of empirical results. We address each major comment below and indicate planned revisions.

read point-by-point responses

  1. Referee: [Abstract / derivation] Abstract and derivation section: the load-bearing premise that 'the records of two evolutions sharing the same decoherence channels differ only in this drift' is asserted without an explicit Radon-Nikodym derivation or verification for the stochastic master equation trajectories used; if control enters the diffusion coefficients, induces cross-qubit correlations, or produces back-action not reducible to drift shift, the claimed closed-form differentiable KL estimator acquires extra factors and the reported performance numbers (including the IBM Kingston ~16% gain) may not follow directly.

    Authors: We agree that an explicit step-by-step Radon-Nikodym derivation for the SME trajectories would improve clarity. The manuscript applies Girsanov's theorem under the stated premise that control modulates only the drift (consistent with standard Lindblad-form open-system models where the decoherence channels are fixed), but we will expand the derivation section to include the full Radon-Nikodym derivative computation and explicitly state the modeling assumptions (control does not alter diffusion coefficients or introduce unaccounted cross-correlations). This will also include a brief discussion of the conditions under which the closed-form KL holds. We believe the reported gains follow directly from these assumptions for the benchmarks considered. revision: yes

  2. Referee: [Empirical results] Empirical evaluation (IBM Kingston chain and multi-qubit benchmarks): the abstract reports gains without error bars, data-exclusion rules, or explicit confirmation that the KL estimator remains free of post-hoc parameter choices or model-specific adjustments; this undermines assessment of whether the +17 pp to +27 pp robustness improvement is reproducible or an artifact of the unverified drift-only assumption.

    Authors: The full manuscript presents results with error bars computed over multiple independent runs (visible in the figures for all benchmarks, including IBM Kingston). We will revise the abstract to reference the statistical evaluation and error bars shown in the results section. We will also add a short methods paragraph confirming that the KL estimator is obtained directly from the Girsanov formula with no post-hoc tuning or model-specific corrections beyond the fixed noise channels. Data exclusion follows standard outlier removal based on convergence criteria, which will be stated explicitly. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation applies external Girsanov theorem

full rationale

The paper asserts that measurement records of two evolutions sharing decoherence channels differ only in drift, allowing Girsanov's theorem to supply a closed-form differentiable KL estimator. This premise is presented as a physical property of the stochastic master equation rather than a self-definition, fitted parameter renamed as prediction, or result derived from self-citation. The two regularizers (KL_W and R_DV) are instantiated from physically motivated reference measures, and performance claims are empirical evaluations on benchmarks. No load-bearing step reduces by construction to the paper's own inputs or prior author work; the derivation remains self-contained against external mathematical results.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the applicability of Girsanov's theorem to continuous-monitoring records of open quantum systems and on the existence of physically motivated reference measures that make the resulting regularizers effective across noise models.

axioms (1)
  • domain assumption Measurement records of two controlled evolutions that share the same decoherence channels differ only in their drift term.
    This premise is required for Girsanov's theorem to yield a closed-form KL estimator between trajectory distributions.

pith-pipeline@v0.9.1-grok · 5828 in / 1376 out tokens · 23841 ms · 2026-06-26T17:33:57.721937+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

70 extracted references · 55 canonical work pages · 5 internal anchors

  1. [1]

    2013 , publisher=

    Quantum error correction , author=. 2013 , publisher=

    2013

  2. [2]

    and Boscain, Ugo and Calarco, Tommaso and Koch, Christiane P

    Glaser, Steffen J. and Boscain, Ugo and Calarco, Tommaso and Koch, Christiane P. and Köckenberger, Walter and Kosloff, Ronnie and Kuprov, Ilya and Luy, Burkhard and Schirmer, Sophie and Schulte-Herbrüggen, Thomas and Sugny, Dominique and Wilhelm, Frank K. , year =. Training. doi:10.1140/epjd/e2015-60464-1 , url =

    work page doi:10.1140/epjd/e2015-60464-1

  3. [3]

    and Boscain, Ugo and Calarco, Tommaso and Dirr, Gunther and Filipp, Stefan and Glaser, Steffen J

    Koch, Christiane P. and Boscain, Ugo and Calarco, Tommaso and Dirr, Gunther and Filipp, Stefan and Glaser, Steffen J. and Kosloff, Ronnie and Montangero, Simone and Schulte-Herbrüggen, Thomas and Sugny, Dominique and Wilhelm, Frank K. , year =. Quantum Optimal Control in Quantum Technologies. doi:10.1140/epjqt/s40507-022-00138-x , url =

    work page doi:10.1140/epjqt/s40507-022-00138-x

  4. [4]

    Quantum sensing , author =. Rev. Mod. Phys. , volume =. 2017 , month =. doi:10.1103/RevModPhys.89.035002 , url =

    work page internal anchor Pith review doi:10.1103/revmodphys.89.035002 2017

  5. [5]

    , year =

    Khaneja, Navin and Reiss, Timo and Kehlet, Cindie and Schulte-Herbrüggen, Thomas and Glaser, Steffen J. , year =. Optimal Control of Coupled Spin Dynamics: Design of. doi:10.1016/j.jmr.2004.11.004 , url =

    work page doi:10.1016/j.jmr.2004.11.004 2004

  6. [6]

    and Ndong, Mamadou and Koch, Christiane P

    Reich, Daniel M. and Ndong, Mamadou and Koch, Christiane P. , year =. Monotonically Convergent Optimization in Quantum Control Using. doi:10.1063/1.3691827 , url =. https://pubs.aip.org/aip/jcp/article-pdf/doi/10.1063/1.3691827/13032598/104103_1_online.pdf , pages =

    work page doi:10.1063/1.3691827

  7. [7]

    Proceedings of the 42nd International Conference on Machine Learning , year=

    Reinforcement Learning for Quantum Control under Physical Constraints , author=. Proceedings of the 42nd International Conference on Machine Learning , year=

  8. [8]

    arXiv preprint arXiv:2410.18635 , year=

    Stochastic optimal control of open quantum systems , author=. arXiv preprint arXiv:2410.18635 , year=

    Pith/arXiv arXiv

  9. [9]

    Girsanov, I. V. , title =. Theory of Probability & Its Applications , volume =. 1960 , doi =

    1960

  10. [10]

    2003 , publisher=

    Stochastic Differential Equations: An Introduction with Applications , author=. 2003 , publisher=

    2003

  11. [11]

    Principle of Minimum Entropy Production , author =. Phys. Rev. , volume =. 1954 , month =. doi:10.1103/PhysRev.96.250 , url =

    work page doi:10.1103/physrev.96.250 1954

  12. [12]

    Principles of maximum entropy and maximum caliber in statistical physics , author =. Rev. Mod. Phys. , volume =. 2013 , month =. doi:10.1103/RevModPhys.85.1115 , url =

    work page doi:10.1103/revmodphys.85.1115 2013

  13. [13]

    2009 , publisher=

    Quantum Measurement and Control , author=. 2009 , publisher=

    2009

  14. [14]

    1998 , publisher=

    Quantum State Diffusion , author=. 1998 , publisher=

    1998

  15. [15]

    Quantum Thermodynamics under Continuous Monitoring:

    Manzano, Gonzalo and Zambrini, Roberta , year =. Quantum Thermodynamics under Continuous Monitoring:. doi:10.1116/5.0079886 , url =. https://pubs.aip.org/avs/aqs/article-pdf/doi/10.1116/5.0079886/19803548/025302_1_online.pdf , pages =

    work page doi:10.1116/5.0079886

  16. [16]

    Advances in Neural Information Processing Systems , author =

    Linearly-Solvable. Advances in Neural Information Processing Systems , author =

  17. [17]

    Linear Theory for Control of Nonlinear Stochastic Systems , author =. Phys. Rev. Lett. , volume =. 2005 , month =. doi:10.1103/PhysRevLett.95.200201 , url =

    work page doi:10.1103/physrevlett.95.200201 2005

  18. [18]

    Thermodynamic Metrics and Optimal Paths , author =. Phys. Rev. Lett. , volume =. 2012 , month =. doi:10.1103/PhysRevLett.108.190602 , url =

    work page doi:10.1103/physrevlett.108.190602 2012

  19. [19]

    Reports on Progress in Physics75(12), 126001 (2012) https://doi.org/ 10.1088/0034-4885/75/12/126001

    Stochastic Thermodynamics, Fluctuation Theorems and Molecular Machines , author =. doi:10.1088/0034-4885/75/12/126001 , url =

    work page doi:10.1088/0034-4885/75/12/126001

  20. [20]

    Quantum-trajectory approach to the stochastic thermodynamics of a forced harmonic oscillator , author =. Phys. Rev. E , volume =. 2012 , month =. doi:10.1103/PhysRevE.85.031110 , url =

    work page doi:10.1103/physreve.85.031110 2012

  21. [21]

    Resonator reset in circuit QED by optimal control for large open quantum systems , author =. Phys. Rev. A , volume =. 2017 , month =. doi:10.1103/PhysRevA.96.042315 , url =

    work page doi:10.1103/physreva.96.042315 2017

  22. [22]

    and Agozzino, Luca and Dill, Ken A

    Ghosh, Kingshuk and Dixit, Purushottam D. and Agozzino, Luca and Dill, Ken A. , year =. The. doi:10.1146/annurev-physchem-071119-040206 , url =

    work page doi:10.1146/annurev-physchem-071119-040206

  23. [23]

    Maximum Caliber: a general variational principle for dynamical systems

    Perspective: Maximum caliber is a general variational principle for dynamical systems. The Journal of Chemical Physics , keywords =. doi:10.1063/1.5012990 , archivePrefix =. 1711.03450 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1063/1.5012990

  24. [24]

    Optimal Control in Large Open Quantum Systems: The Case of Transmon Readout and Reset , author =. Phys. Rev. Lett. , volume =. 2025 , month =. doi:10.1103/PhysRevLett.134.070802 , url =

    work page doi:10.1103/physrevlett.134.070802 2025

  25. [25]

    , journal =

    Karmakar, Tathagata and Jordan, Andrew N. , journal =. doi:10.22331/q-2026-03-24-2043 , url =

    work page doi:10.22331/q-2026-03-24-2043 2026

  26. [26]

    Faidon Brotzakis and Michele Vendruscolo and Peter G

    Z. Faidon Brotzakis and Michele Vendruscolo and Peter G. Bolhuis , title =. Proceedings of the National Academy of Sciences , volume =. 2021 , doi =. https://www.pnas.org/doi/pdf/10.1073/pnas.2012423118 , abstract =

    work page doi:10.1073/pnas.2012423118 2021

  27. [27]

    2021 , journaltitle =

    A Maximum Caliber Approach for Continuum Path Ensembles , author =. 2021 , journaltitle =. doi:10.1140/epjb/s10051-021-00154-3 , url =

    work page doi:10.1140/epjb/s10051-021-00154-3 2021

  28. [28]

    Gradient-based optimal control of open quantum systems using quantum trajectories and automatic differentiation , author =. Phys. Rev. A , volume =. 2019 , month =. doi:10.1103/PhysRevA.99.052327 , url =

    work page doi:10.1103/physreva.99.052327 2019

  29. [29]

    2023 , journaltitle =

    Optimizing Molecular Potential Models by Imposing Kinetic Constraints with Path Reweighting , author =. 2023 , journaltitle =. doi:10.1063/5.0151166 , url =. https://pubs.aip.org/aip/jcp/article-pdf/doi/10.1063/5.0151166/18299649/074102_1_5.0151166.pdf , pages =

    work page doi:10.1063/5.0151166 2023

  30. [30]

    arXiv preprint arXiv:2509.24104 , year=

    Path integral quantum control for quantum chemistry applications , author=. arXiv preprint arXiv:2509.24104 , year=

    arXiv

  31. [31]

    Preskill, John , journal =. Quantum. doi:10.22331/q-2018-08-06-79 , url =

    work page internal anchor Pith review doi:10.22331/q-2018-08-06-79 2018

  32. [32]

    Decoherence, einselection, and the quantum origins of the classical , author =. Rev. Mod. Phys. , volume =. 2003 , month =. doi:10.1103/RevModPhys.75.715 , url =

    work page doi:10.1103/revmodphys.75.715 2003

  33. [33]

    Decoherence-Free Subspaces for Quantum Computation , author =. Phys. Rev. Lett. , volume =. 1998 , month =. doi:10.1103/PhysRevLett.81.2594 , url =

    work page doi:10.1103/physrevlett.81.2594 1998

  34. [34]

    Dynamical Decoupling of Open Quantum Systems , author =. Phys. Rev. Lett. , volume =. 1999 , month =. doi:10.1103/PhysRevLett.82.2417 , url =

    work page internal anchor Pith review doi:10.1103/physrevlett.82.2417 1999

  35. [35]

    Lindblad

    On the Generators of Quantum Dynamical Semigroups , author =. 1976 , date =. doi:10.1007/BF01608499 , url =

    work page doi:10.1007/bf01608499 1976

  36. [36]

    Universal Fidelity Reduction of Quantum Operations from Weak Dissipation , author =. Phys. Rev. Lett. , volume =. 2022 , month =. doi:10.1103/PhysRevLett.129.150504 , url =

    work page doi:10.1103/physrevlett.129.150504 2022

  37. [37]

    Gorini, Vittorio and Kossakowski, Andrzej and Sudarshan, E. C. G. , year =. Completely Positive Dynamical Semigroups of. doi:10.1063/1.522979 , url =. https://pubs.aip.org/aip/jmp/article-pdf/17/5/821/19090720/821_1_online.pdf , pages =

    work page doi:10.1063/1.522979

  38. [38]

    2021 , journaltitle =

    Leakage Reduction in Fast Superconducting Qubit Gates via Optimal Control , author =. 2021 , journaltitle =. doi:10.1038/s41534-020-00346-2 , url =

    work page doi:10.1038/s41534-020-00346-2 2021

  39. [39]

    Simple Pulses for Elimination of Leakage in Weakly Nonlinear Qubits , author =. Phys. Rev. Lett. , volume =. 2009 , month =. doi:10.1103/PhysRevLett.103.110501 , url =

    work page doi:10.1103/physrevlett.103.110501 2009

  40. [40]

    Quantum metrology with nonclassical states of atomic ensembles , author =. Rev. Mod. Phys. , volume =. 2018 , month =. doi:10.1103/RevModPhys.90.035005 , url =

    work page internal anchor Pith review doi:10.1103/revmodphys.90.035005 2018

  41. [41]

    B.et al.From pulses to circuits and back again: A quantum optimal control perspective 11 on variational quantum algorithms.PRX Quantum 2, 010101 (2021)

    From Pulses to Circuits and Back Again: A Quantum Optimal Control Perspective on Variational Quantum Algorithms , author =. PRX Quantum , volume =. 2021 , month =. doi:10.1103/PRXQuantum.2.010101 , url =

    work page doi:10.1103/prxquantum.2.010101 2021

  42. [42]

    Kappen, H. J. and Ruiz, H. C. , year =. Adaptive. doi:10.1007/s10955-016-1446-7 , url =

    work page doi:10.1007/s10955-016-1446-7

  43. [43]

    Advances in Neural Information Processing Systems , volume=

    Stochastic optimal control for collective variable free sampling of molecular transition paths , author=. Advances in Neural Information Processing Systems , volume=

  44. [44]

    Orlando, Simon Gustavsson, and William D

    A Quantum Engineer's Guide to Superconducting Qubits , author =. 2019 , journaltitle =. doi:10.1063/1.5089550 , url =. https://pubs.aip.org/aip/apr/article-pdf/doi/10.1063/1.5089550/20722375/021318_1_1.5089550.pdf , pages =

    work page doi:10.1063/1.5089550 2019

  45. [45]

    arXiv preprint arXiv:2506.15420 , year=

    Error-detected coherence metrology of a dual-rail encoded fixed-frequency multimode superconducting qubit , author=. arXiv preprint arXiv:2506.15420 , year=

    arXiv

  46. [46]

    Erasure Qubits: Overcoming the

    Kubica, Aleksander and Haim, Arbel and Vaknin, Yotam and Levine, Harry and Brand\ ao, Fernando and Retzker, Alex , journal =. Erasure Qubits: Overcoming the. 2023 , month =. doi:10.1103/PhysRevX.13.041022 , url =

    work page doi:10.1103/physrevx.13.041022 2023

  47. [47]

    Stimulated Raman adiabatic passage in physics, chemistry, and beyond , author =. Rev. Mod. Phys. , volume =. 2017 , month =. doi:10.1103/RevModPhys.89.015006 , url =

    work page doi:10.1103/revmodphys.89.015006 2017

  48. [48]

    Roadmap on

    Bergmann, Klaas and Nägerl, Hanns-Christoph and Panda, Cristian and Gabrielse, Gerald and Miloglyadov, Eduard and Quack, Martin and Seyfang, Georg and Wichmann, Gunther and Ospelkaus, Silke and Kuhn, Axel and Longhi, Stefano and Szameit, Alexander and Pirro, Philipp and Hillebrands, Burkard and Zhu, Xue-Feng and Zhu, Jie and Drewsen, Michael and Hensinger...

    work page doi:10.1088/1361-6455/ab3995

  49. [49]

    doi:10.1038/nature10401 , url =

    High-Fidelity Projective Read-out of a Solid-State Spin Quantum Register , author =. doi:10.1038/nature10401 , url =

    work page doi:10.1038/nature10401

  50. [50]

    and Chakram, S

    Earnest, N. and Chakram, S. and Lu, Y. and Irons, N. and Naik, R. K. and Leung, N. and Ocola, L. and Czaplewski, D. A. and Baker, B. and Lawrence, Jay and Koch, Jens and Schuster, D. I. , journal =. Realization of a. 2018 , month =. doi:10.1103/PhysRevLett.120.150504 , url =

    work page doi:10.1103/physrevlett.120.150504 2018

  51. [51]

    doi:10.1038/s41467-021-26686-x , url =

    Electron Shelving of a Superconducting Artificial Atom , author =. doi:10.1038/s41467-021-26686-x , url =

    work page doi:10.1038/s41467-021-26686-x

  52. [52]

    Quantum dynamics of single trapped ions , author =. Rev. Mod. Phys. , volume =. 2003 , month =. doi:10.1103/RevModPhys.75.281 , url =

    work page doi:10.1103/revmodphys.75.281 2003

  53. [53]

    Quantum Communication through an Unmodulated Spin Chain , author =. Phys. Rev. Lett. , volume =. 2003 , month =. doi:10.1103/PhysRevLett.91.207901 , url =

    work page doi:10.1103/physrevlett.91.207901 2003

  54. [54]

    Transfer of arbitrary two-qubit states via a spin chain , author =. Phys. Rev. A , volume =. 2015 , month =. doi:10.1103/PhysRevA.91.042321 , url =

    work page doi:10.1103/physreva.91.042321 2015

  55. [55]

    2005 , journaltitle =

    Quantum State Transfer in Arrays of Flux Qubits , author =. 2005 , journaltitle =. doi:10.1088/1367-2630/7/1/181 , url =

    work page doi:10.1088/1367-2630/7/1/181 2005

  56. [56]

    doi:10.1038/s41534-019-0168-5 , url =

    Decoherence Benchmarking of Superconducting Qubits , author =. doi:10.1038/s41534-019-0168-5 , url =

    work page doi:10.1038/s41534-019-0168-5

  57. [57]

    doi:10.1038/s41534-021-00448-5 , url =

    Time-Varying Quantum Channel Models for Superconducting Qubits , author =. doi:10.1038/s41534-021-00448-5 , url =

    work page doi:10.1038/s41534-021-00448-5

  58. [58]

    International conference on machine learning , pages=

    Mutual information neural estimation , author=. International conference on machine learning , pages=. 2018 , organization=

    2018

  59. [59]

    2021 , journaltitle =

    Omg Blueprint for Trapped Ion Quantum Computing with Metastable States , author =. 2021 , journaltitle =. doi:10.1063/5.0069544 , url =. https://pubs.aip.org/aip/apl/article-pdf/doi/10.1063/5.0069544/19773783/214002_1_online.pdf , pages =

    work page doi:10.1063/5.0069544 2021

  60. [60]

    2022 , journaltitle =

    Realizing Coherently Convertible Dual-Type Qubits with the Same Ion Species , author =. 2022 , journaltitle =. doi:10.1038/s41567-022-01661-5 , url =

    work page doi:10.1038/s41567-022-01661-5 2022

  61. [61]

    , year =

    Wu, Yue and Kolkowitz, Shimon and Puri, Shruti and Thompson, Jeff D. , year =. Erasure Conversion for Fault-Tolerant Quantum Computing in Alkaline Earth. doi:10.1038/s41467-022-32094-6 , url =

    work page doi:10.1038/s41467-022-32094-6

  62. [62]

    Demonstrating a Long-Coherence Dual-Rail Erasure Qubit Using Tunable Transmons , author =. Phys. Rev. X , volume =. 2024 , month =. doi:10.1103/PhysRevX.14.011051 , url =

    work page doi:10.1103/physrevx.14.011051 2024

  63. [63]

    Ivanov, P. A. and Poschinger, U. G. and Singer, K. and Schmidt-Kaler, F. , title =. Europhysics Letters , abstract =. 2010 , month =. doi:10.1209/0295-5075/92/30006 , url =

    work page doi:10.1209/0295-5075/92/30006 2010

  64. [64]

    Fast and Unconditional All-Microwave Reset of a Superconducting Qubit , author =. Phys. Rev. Lett. , volume =. 2018 , month =. doi:10.1103/PhysRevLett.121.060502 , url =

    work page doi:10.1103/physrevlett.121.060502 2018

  65. [65]

    Pulsed Reset Protocol for Fixed-Frequency Superconducting Qubits , author =. Phys. Rev. Appl. , volume =. 2018 , month =. doi:10.1103/PhysRevApplied.10.044030 , url =

    work page doi:10.1103/physrevapplied.10.044030 2018

  66. [66]

    Spontaneous emission in stimulated Raman adiabatic passage , author =. Phys. Rev. A , volume =. 2005 , month =. doi:10.1103/PhysRevA.72.053412 , url =

    work page doi:10.1103/physreva.72.053412 2005

  67. [67]

    High-fidelity entanglement of metastable trapped-ion qubits with integrated erasure conversion , author =. Phys. Rev. A , volume =. 2026 , month =. doi:10.1103/p3cy-8yjk , url =

    work page doi:10.1103/p3cy-8yjk 2026

  68. [68]

    Annual Review of Physical Chemistry , volume=

    The minimum entropy production principle , author=. Annual Review of Physical Chemistry , volume=. 1980 , publisher=

    1980

  69. [69]

    Complex Systems—Operational Approaches in Neurobiology, Physics, and Computers: Proceedings of the International Symposium on Synergetics at Schlo

    Macroscopic prediction , author=. Complex Systems—Operational Approaches in Neurobiology, Physics, and Computers: Proceedings of the International Symposium on Synergetics at Schlo. 1985 , organization=

    1985

  70. [70]

    2025 , url =

    Dynamiqs: an open-source Python library for GPU-accelerated and differentiable simulation of quantum systems , author =. 2025 , url =

    2025