Pith. sign in

REVIEW 3 major objections 5 minor 72 references

Trading Imaginary Time for Randomness in Ground State Preparation

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Twirled imaginary-time evolution replaces half the imaginary time with random real-time kicks, quadratically suppressing ground-state error.

desk verdict TITE is a genuinely new and mostly rigorous way to halve the imaginary-time cost of ground state preparation; the one real gap is the unproven real-time Trotter error budget, which a referee should push on. read the letter →

arxiv 2608.00443 v1 pith:PBXF7OTW submitted 2026-08-01 quant-ph

Alvan Arulandu , John M. Martyn , Isaac L. Chuang This is my paper · ORCID
classification quant-ph
keywords imaginary-timeevolutiongroundstatepreparationrandomizedcompilingtemporaltwirlingtracedistancequantumsimulationmixinglemmaerrormitigation
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

The paper introduces twirled imaginary-time evolution (TITE), which pairs imaginary-time evolution with real-time evolution for a random duration drawn from a carefully designed distribution. It proves that this randomization suppresses the trace distance to the ground state quadratically, from $O(\epsilon)$ to $O(\epsilon^2)$, so that the imaginary time needed for a fixed accuracy drops from $\beta$ to about $\beta/2$. Because standard implementations of imaginary-time evolution incur costs that scale exponentially in $\beta$, replacing half the imaginary time with unitary real-time evolution yields a quadratic reduction in cost for any black-box ITE method. The paper demonstrates the effect on a non-integrable Ising chain, showing doubled error-decay rates in noisy circuit-level and noiseless simulations.

What carries the argument

The load-bearing object is the mixing lemma for states (Lemma 2), a state-level analogue of the Campbell-Hastings mixing lemma: if an ensemble of pure states $\{|u_j\rangle\}$ is individually $a$-close to a target $|v\rangle$ and their average is $b$-close, then the mixed state $\mathbb{E}_j[|u_j\rangle\langle u_j|]$ is at trace distance $a^2/2 + b$ from $|v\rangle\langle v|$. TITE instantiates it with $|u_t\rangle = e^{-iHt}|\psi(\beta)\rangle$ and chooses $D$ so that $|\mathbb{E}_t[e^{-i\omega t}]| \le O(\epsilon)$ for all $\omega \ge \Delta$, the spectral-gap bound. The paper constructs three such distributions: $S_\Delta$ with density $\frac{3\Delta}{8\pi}\operatorname{sinc}^4(\Delta t/4)$, whose characteristic function vanishes identically for $|\omega| \ge \Delta$ and whose moments are optimal; $C_\beta$ with density $\frac{1}{\beta\pi(1+(t/\beta)^4/4)}$, for when $\Delta$ is unknown; and a Gaussian $N(0, 2\beta/\Delta)$ that is suboptimal but standard. The temporal twirling channel $\rho \mapsto \mathbb{E}_t[e^{-iHt}\rho e^{iHt}]$ is the mechanism that turns coherent excited-state error into incoherent error.

What would settle it

Measure the total circuit cost needed to reach a fixed trace distance $\epsilon$ as $\beta$ grows, while demanding real-time Trotter error at most $O(\epsilon^2)$. If the number of real-time Trotter steps must grow exponentially in $\beta$ to maintain the $e^{-2\beta\Delta}$ decay, the halved imaginary time is offset and the claimed quadratic reduction in total cost fails.

Watch

Extended reading notes

Core claim

The central discovery is that randomness can substitute for imaginary time in ground-state preparation. Given a state $|\psi(\beta)\rangle$ obtained by imaginary-time evolution that is $O(\epsilon)$-close to the ground state, applying real-time evolution $e^{-iHt}$ for $t$ drawn from a distribution $D$ whose characteristic function decays to $O(\epsilon)$ for all frequencies above the spectral gap renders the excited-state coherences incoherent. The resulting mixed state is $O(\epsilon^2)$-close to the ground state in trace distance. Equivalently, TITE achieves the same accuracy as ITE with $\beta \to \beta/2$, and since the cost of black-box ITE implementations scales as $e^{O(\beta)}$, this is a quadratic cost reduction. The paper also proves this quadratic suppression is optimal within the class of unitaries that stabilize the ground state up to phase.

Load-bearing premise

The practical advantage assumes that real-time evolution can be implemented with Trotter error $O(\epsilon^2)$ without incurring a cost that cancels the savings from halving $\beta$; the paper states this can be seen but does not prove it, and the numerics fix the real-time Trotter step count rather than scaling it.

Editorial extensions

If this is right

  • Any ITE implementation whose cost scales as $e^{O(\beta)}$ — Trotterization with post-selection, quantum imaginary-time evolution, or quantum signal processing — inherits a quadratic reduction in cost, because $\beta$ can be halved without losing accuracy.
  • For target accuracy $\epsilon$, TITE requires $\beta = O(\log(1/\epsilon)/\Delta)$ rather than twice that, and the state error, not just the energy error, is suppressed; arbitrary observables benefit from the $O(\epsilon^2)$ trace-distance bound.
  • The advantage persists under moderate depolarizing gate noise and under Trotterization of both the imaginary- and real-time evolution, provided each Trotter error stays below $O(\epsilon^2)$.
  • The quadratic suppression cannot be improved within this method: any twirling with unitaries that stabilize the ground state up to phase leaves a $\delta^2$ floor in trace distance (Lemma 4).

Reading between the lines

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

  • Because temporal twirling is a channel-level operation, it could be layered on top of other ground-state preparation methods, including variational or dissipative approaches, without adding variational parameters; the paper mentions this direction but does not analyze it in detail.
  • A testable prediction is that TITE's advantage grows for observables that are off-diagonal in the energy eigenbasis and shrinks for observables nearly diagonal in it; the numerical magnetization data hint at this ordering but no general statement is proven.
  • If the clock-register purification of the randomness is measured and post-selected rather than traced out, the resulting spectral filter could break the $\delta^2$ floor at the cost of sample complexity; quantifying that trade-off would be a natural extension.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper introduces twirled imaginary-time evolution (TITE), a method that augments imaginary-time evolution (ITE) for ground-state preparation with real-time evolution applied for a random duration t drawn from a distribution D. The main theoretical result, Theorem 3, states that if ITE for time β brings the state within trace distance O(ε) of the ground state, and D is chosen so that its characteristic function satisfies |E_D[e^{-itω}]| ≤ O(ε) for all ω ≥ Δ, then the randomized real-time evolution produces a mixed state within trace distance O(ε²) of the ground state. This quadratic suppression is proved via a mixing lemma for states (Lemma 2). The paper constructs three distributions—S_Δ, C_β, and N_{β,Δ}—with explicit characteristic functions, moment bounds, and optimality claims, and reports noiseless and noisy circuit-level simulations on a 10-site non-integrable Ising chain that show the predicted doubled decay rate of the trace distance. The authors further argue that the quadratic suppression lets half of the imaginary time be replaced by real-time evolution, yielding a quadratic reduction in the e^{O(β)} cost of black-box ITE implementations.

Significance. The core mathematical observation is elegant and potentially useful: randomizing over real-time evolution can convert first-order coherent error in an approximate ground state into second-order incoherent error, without changing the state populations. Theorem 3 and Lemma 2 are proved in full, and Appendix B gives rigorous, self-contained constructions of distributions with the required characteristic-function decay and moment scaling, including optimality lower bounds. The numerical experiments are a strength: the authors simulate the actual circuits with depolarizing noise, isolate Trotterization effects in noiseless runs, and compare all three distributions, with code and data made openly available. If the end-to-end cost claim can be substantiated, the result would be a broadly applicable reduction for ITE-based ground-state preparation. However, the paper's central practical claim—quadratic reduction in cost—currently rests on an unproven assumption about the cost of implementing the real-time evolution with error O(ε²), and on a missing comparison between that cost and the saved imaginary-time sample cost.

major comments (3)
  1. [Section III.D and Figure 6] The assertion that real-time evolution (RTE) error O(ε²) 'straightforwardly' preserves the TITE error suppression is not proven, and it is load-bearing for the central claim. Theorem 3 is stated for exact e^{-iHt}; if each RTE call is replaced by a Trotterized channel with diamond-norm error η, the triangle inequality gives d_tr(ρ, |λ0⟩⟨λ0|) ≤ O(ε²) + η, so maintaining the O(ε²) suppression requires η = O(ε²). The manuscript does not state this as a lemma, does not specify how the number of Trotter steps r_RTE must scale with β, ε, Δ, and t, and does not show that the Trotter error satisfies the conditions a=O(ε), b=O(ε²) assumed in Lemma 2. Figure 6 shows that with the fixed r_RTE=100 used in the simulations, Trotter error dominates beyond a critical β_c^RTE and produces a plateau. A quantitative analysis of this Trotter-error budget is needed to substantiate the claim that the quadratic suppression survives under a concrete RTE implementation.
  2. [Section III.D and Abstract] The claimed 'quadratic reduction in the cost of any black-box ITE implementation' is an end-to-end resource statement, but the paper compares only the imaginary-time sample cost before and after β→β/2. Real-time evolution is unitary and does not cost samples or classical post-processing, but it still consumes quantum gates; that cost must be counted. For example, with first-order Trotterization one has C_RTE(t,ε)=O(t²/ε) for RTE error ε, and for the optimal distribution S_Δ the second moment is E[t²]=12/Δ², giving an expected RTE overhead O(1/(Δ² ε²)) per shot to reach RTE error O(ε²). The sample-complexity saving from halving β is e^{β/2}=(c_min ε)^{-1/(2Δ)} up to constants. When Δ>1/4, ε^{-1/(2Δ)} grows more slowly than ε^{-2}, so the RTE gate cost dominates asymptotically and the claimed quadratic reduction does not follow. The paper should either identify the parameter regime where the reduction holds, invoke a near-optimal Hamiltonian simulation subroutine with polylog(1/ε) cost for the RTE step, or explicitly restrict the claim to sample/classical cost and list the RTE gate cost as a separate resource.
  3. [Section III.D and Section V] The paper says in Section V that a complete noise analysis remains future work, but the numerical advantage in the noisy regime is part of the paper's evidence for practical utility. Figure 5 shows that the TITE advantage over ITE shrinks as the depolarizing noise strength increases, because the additional real-time evolution circuits are deeper. The manuscript does not quantify the crossover noise strength at which the advantage disappears, nor does it provide any argument that the noise incurred by the RTE gadgets is comparable to or less than the noise saved by halving the imaginary-time circuit depth. Since the motivation in the introduction emphasizes near-term and early fault-tolerant implementations, this missing analysis should at least be flagged as a condition on the validity of the numerical claim in Section IV.C.
minor comments (5)
  1. [Section III.E] The phrase 'artisanal C_β and S_Δ distributions' in Section IV.C is informal; 'purpose-built' or 'tailored' would be more appropriate for a journal style.
  2. [Figure 4 caption] The term 'open-controlled' is used without definition; define it at first use or use 'controlled on the |0⟩ state' to avoid ambiguity.
  3. [Appendix B, Lemma 5] In the moment computation, the displayed final expression has a removable singularity at α=3, and the text states that the moment diverges logarithmically there; a brief derivation of the logarithmic divergence would make the statement self-contained.
  4. [Data Availability] Reference [68] gives only '[github]' with no URL or DOI; for reproducibility, include a persistent identifier or a full URL.
  5. [Section III.C] The remark that the method 'also applies to any gapped excited state' should state that the distribution condition of Eq. (14) must then hold for all relevant transition frequencies relative to that excited state, not just for ω≥Δ as written for the ground state.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: TITE's quadratic suppression theorem and distribution constructions are self-contained, and the self-citations are not load-bearing.

full rationale

The central derivation (Theorem 3 and Sec. III.E) is mathematically self-contained: the trace-distance suppression is proved from the state mixing lemma using only the characteristic-function condition |E_D[e^{-iωt}]| ≤ O(ε) for ω ≥ Δ, and the distributions S_Δ, C_β, and N_{β,Δ} are constructed and optimized in Appendix B rather than fitted to data. The numerical experiments simulate the actual circuits without fitting any parameter to the theory, so the quadratic advantage is not a fitted input renamed as a prediction. The self-citations (Refs. [53] and [65], sharing author Martyn) appear only as background examples and future-work pointers, not as premises of Theorem 3 or of the cost-reduction claim. The statement in Sec. III.D that real-time Trotter error O(ε²) preserves the suppression is asserted without proof and is a correctness or rigor gap rather than a circular step; likewise Fig. 6 shows a Trotter-error plateau, but this is an implementation caveat, not a reduction of the theorem to its assumptions.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central theorem rests on the gapped non-degenerate Hamiltonian and positive overlap assumptions, plus the exponential-cost model for ITE implementations. The distributions used for twirling are constructed with proofs, so they are not ad hoc free parameters. No new physical entities are introduced.

assumptions (4)
  • domain assumption The Hamiltonian has a gapped, non-degenerate ground state |λ0⟩ with spectral gap lower bound Δ > 0.
    Problem 1 and Theorem 3 assume this. The trace distance analysis relies on exponential suppression of excited states at rate Δ.
  • domain assumption The initial state has non-zero ground state overlap |⟨λ0|ψ⟩| ≥ c_min > 0.
    Required for the normalization and for β = log(1/(c_min ε))/Δ to yield δ ≤ ε in Theorem 3.
  • domain assumption The black-box ITE implementation has sample/classical cost scaling as e^{O(β)}.
    This is the premise on which the 'quadratic reduction in cost' claim rests; it is argued from the literature (post-selection success probability, QITE domain growth) and assumed in the complexity analysis.
  • standard math Standard Fourier analysis identities (e.g., from Gradshteyn and Ryzhik) used in Appendix B.
    Used to compute moments and characteristic functions of SΔ, Cβ, Nβ,Δ.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Trading Imaginary Time for Randomness in Ground State Preparation." pith.science (2026). https://pith.science/paper/PBXF7OTW

@misc{pith2026260800443,
  author       = {Pith},
  title        = {Pith review of: Trading Imaginary Time for Randomness in Ground State Preparation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PBXF7OTW}},
  note         = {Machine review of arXiv:2608.00443}
}
abstract

Imaginary-time evolution (ITE) is a foundational method for ground state preparation on quantum computers. However, because ITE is non-unitary, existing implementations incur a sample complexity and/or classical cost that scales exponentially with the target imaginary time $\beta$. Moreover, the state itself converges slower than the energy, making accurate estimation of arbitrary ground state observables even more expensive. In this work, we improve upon standard ITE by introducing twirled imaginary-time evolution (TITE), which pairs ITE with real-time evolution applied for a random duration drawn from a carefully designed distribution. We prove that this randomization quadratically suppresses the trace distance to the ground state, and thus also the error of arbitrary observables, which allows roughly half of the imaginary time to be replaced with real-time evolution ($\beta \mapsto \beta/2$) while maintaining the same level of accuracy. Because real-time evolution is unitary and does not incur an overhead in sample complexity or classical computation, this affords a quadratic reduction in the cost of any black-box ITE implementation, including Trotterization and quantum imaginary-time evolution. We demonstrate the efficiency of our algorithm in noisy circuit-level simulations of a non-integrable Ising chain, showing substantial improvements over standard ITE.

Figures

Figures reproduced from arXiv: 2608.00443 by the authors.

Figure 1
Figure 1. The Campbell-Hastings mixing lemma (Lemma [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Illustration of the twirled imaginary time evolu [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Schematic comparison (not to scale) of the char [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Circuit gadgets for implementing real-time evolu [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 5
Figure 5. Figure 5: Noisy circuit simulation (n = 10), comparing ITE to TITE (D = Cβ) across noise strengths γ ∈ {0, 10−6 , 10−5 }. twirled method worsens yet is still considerable, though the baseline ITE is more noise-robust. This is because the random real-time evolution increases the …
Figure 8
Figure 8. Figure 8: Noiseless simulation of the population error and co [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]
Figure 7
Figure 7. Figure 7: Noiseless comparison of the three twirling distribu [PITH_FULL_IMAGE:figures/full_fig_p012_7.png]
Figure 9
Figure 9. Figure 9: Optimality of quadratic error suppression [PITH_FULL_IMAGE:figures/full_fig_p014_9.png]
Figure 10
Figure 10. Figure 10: The three twirling distributions of this appendix, plotted exactly with ∆ = 1 and [PITH_FULL_IMAGE:figures/full_fig_p016_10.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

72 extracted references · 37 canonical work pages

  1. [1]

    Then, for any basic structured ensembleA, |⟨A,W⟩|=|⟨x|E j[Wj]|y⟩|≤∥E j[Wj]∥op≤b where the final inequality is precisely the mean condi- tion of the mixing lemma

    bound both the individual magnitudes of the errors, ∥Wj∥op =∥U j−V∥ op≤a, as well as the mean of the errors,∥E j[Wj]∥op =∥ P jpjUj−V∥ op≤b. Then, for any basic structured ensembleA, |⟨A,W⟩|=|⟨x|E j[Wj]|y⟩|≤∥E j[Wj]∥op≤b where the final inequality is precisely the mean condi- tion of the mixing lemma. This is exactly the statement thatWisb-pseudorandom [60...

    work page

  2. [2]

    WhileS ∆ andC β are optimal in their own respects, they are both purpose-built

    Repeating the two estimates in the proof of Lemma 6 with ∆ replaced by a/βandcϵreplaced by 1 2 yields ED[|t|]≥ β 2a,E D[t2]≥ β2 a2, and the claim forα∈[1,2] follows by Jensen’s inequality. WhileS ∆ andC β are optimal in their own respects, they are both purpose-built. We conclude by showing that the desired decay constraints can be achieved by more common...

    work page

  3. [3]

    T. I. Andersen, N. Astrakhantsev, A. H. Karamlou, J. Berndtsson, J. Motruk, A. Szasz, J. A. Gross, A. Schuckert, T. Westerhout, Y. Zhang,et al., Thermal- ization and criticality on an analogue–digital quantum simulator, Nature638, 79 (2025)

    2025

  4. [4]

    Y. Kim, A. Eddins, S. Anand, K. X. Wei, E. Van Den Berg, S. Rosenblatt, H. Nayfeh, Y. Wu, M. Zale- tel, K. Temme,et al., Evidence for the utility of quan- tum computing before fault tolerance, Nature618, 500 (2023)

    2023

  5. [5]

    Y. Cao, J. Romero, J. P. Olson, M. Degroote, P. D. John- son, M. Kieferov´ a, I. D. Kivlichan, T. Menke, B. Per- opadre, N. P. D. Sawaya, S. Sim, L. Veis, and A. Aspuru- Guzik, Quantum chemistry in the age of quantum com- puting, Chemical Reviews119, 10856 (2019)

    2019

  6. [6]

    Preskill, Simulating quantum field theory with a quan- tum computer, inThe 36th Annual International Sym- posium on Lattice Field Theory

    J. Preskill, Simulating quantum field theory with a quan- tum computer, inThe 36th Annual International Sym- posium on Lattice Field Theory. 22-28 July(2018) p. 24, arXiv:1811.10085 [hep-lat]

    arXiv 2018

  7. [7]

    S. P. Jordan, K. S. M. Lee, and J. Preskill, Quantum computation of scattering in scalar quantum field theo- ries, Quantum Info. Comput.14, 1014–1080 (2014)

    work page 2014

  8. [8]

    Lin and Y

    L. Lin and Y. Tong, Near-optimal ground state prepara- tion, Quantum4, 372 (2020)

    2020

Show all 72 references
  1. [9]

    Y. Dong, L. Lin, and Y. Tong, Ground-state prepara- tion and energy estimation on early fault-tolerant quan- tum computers via quantum eigenvalue transformation of unitary matrices, PRX Quantum3, 040305 (2022)

    2022

  2. [10]

    X. Peng, J. Du, and D. Suter, Quantum phase transition of ground-state entanglement in a Heisenberg spin chain simulated in an NMR quantum computer, Phys. Rev. A 71, 012307 (2005)

    2005

  3. [11]

    Cerezo, K

    M. Cerezo, K. Sharma, A. Arrasmith, and P. J. Coles, Variational quantum state eigensolver, npj Quantum In- formation8, 113 (2022)

    2022

  4. [12]

    H. Yu, Y. Zhao, and T.-C. Wei, Simulating large-size quantum spin chains on cloud-based superconducting quantum computers, Phys. Rev. Res.5, 013183 (2023)

    2023

  5. [13]

    Peruzzo, J

    A. Peruzzo, J. McClean, P. Shadbolt, M.-H. Yung, X.-Q. Zhou, P. J. Love, A. Aspuru-Guzik, and J. L. O’Brien, A variational eigenvalue solver on a photonic quantum processor, Nature communications5, 4213 (2014)

    2014

  6. [14]

    P. J. J. O’Malley, R. Babbush, I. D. Kivlichan, J. Romero, J. R. McClean, R. Barends, J. Kelly, P. Roushan, A. Tranter, N. Ding, B. Campbell, Y. Chen, Z. Chen, B. Chiaro, A. Dunsworth, A. G. Fowler, E. Jef- frey, E. Lucero, A. Megrant, J. Y. Mutus, M. Neeley, C. Neill, C. Quin...

    2016

  7. [15]

    Hempel, C

    C. Hempel, C. Maier, J. Romero, J. McClean, T. Monz, H. Shen, P. Jurcevic, B. P. Lanyon, P. Love, R. Babbush, A. Aspuru-Guzik, R. Blatt, and C. F. Roos, Quantum chemistry calculations on a trapped-ion quantum simu- lator, Phys. Rev. X8, 031022 (2018)

    2018

  8. [16]

    Romero, R

    J. Romero, R. Babbush, J. R. McClean, C. Hempel, P. J. Love, and A. Aspuru-Guzik, Strategies for quantum com- puting molecular energies using the unitary coupled clus- ter ansatz, Quantum Science and Technology4, 014008 (2018)

    2018

  9. [17]

    Farhi, J

    E. Farhi, J. Goldstone, and S. Gutmann, A quan- tum approximate optimization algorithm (2014), arXiv:1411.4028 [quant-ph]

    2014 arXiv

  10. [18]

    S. P. Jordan, N. Shutty, M. Wootters, A. Zalcman, A. Schmidhuber, R. King, S. V. Isakov, T. Khattar, and R. Babbush, Optimization by decoded quantum interfer- ometry (2025), arXiv:2408.08292 [quant-ph]

    2025

  11. [19]

    Xie, S.-J

    H.-N. Xie, S.-J. Wei, F. Yang, Z.-A. Wang, C.-T. Chen, H. Fan, and G.-L. Long, Probabilistic imaginary-time evolution algorithm based on nonunitary quantum cir- cuits, Phys. Rev. A109, 052414 (2024)

    2024

  12. [20]

    Kosugi, Y

    T. Kosugi, Y. Nishiya, H. Nishi, and Y.-i. Matsushita, Imaginary-time evolution using forward and backward real-time evolution with a single ancilla: First-quantized eigensolver algorithm for quantum chemistry, Phys. Rev. Res.4, 033121 (2022)

    2022

  13. [21]

    Turro, A

    F. Turro, A. Roggero, V. Amitrano, P. Luchi, K. A. Wendt, J. L. Dubois, S. Quaglioni, and F. Pederiva, Imaginary-time propagation on a quantum chip, Phys. Rev. A105, 022440 (2022)

    2022

  14. [22]

    Liu, J.-G

    T. Liu, J.-G. Liu, and H. Fan, Probabilistic nonunitary gate in imaginary time evolution, Quantum Information Processing20(2021)

    2021

  15. [23]

    Leadbeater, N

    C. Leadbeater, N. Fitzpatrick, D. Mu˜ noz Ramo, and A. J. Thom, Non-unitary Trotter circuits for imagi- nary time evolution, Quantum Science and Technology 9, 045007 (2024)

    2024

  16. [24]

    Motta, C

    M. Motta, C. Sun, A. T. K. Tan, M. J. O’Rourke, E. Ye, A. J. Minnich, F. G. S. L. Brand˜ ao, and G. K.-L. Chan, Determining eigenstates and thermal states on a quan- tum computer using quantum imaginary time evolution, Nature Physics16, 205–210 (2019)

    2019

  17. [25]

    Gily´ en, Y

    A. Gily´ en, Y. Su, G. H. Low, and N. Wiebe, Quantum singular value transformation and beyond: exponential improvements for quantum matrix arithmetics, inPro- ceedings of the 51st Annual ACM SIGACT Symposium on Theory of Computing, STOC ’19 (ACM, 2019)

    2019

  18. [26]

    T. L. Silva, M. M. Taddei, S. Carrazza, and L. Aolita, Fragmented imaginary-time evolution for early-stage quantum signal processors, Scientific Reports13(2023). 19

    2023

  19. [27]

    S.-N. Sun, M. Motta, R. N. Tazhigulov, A. T. Tan, G. K.-L. Chan, and A. J. Minnich, Quantum computa- tion of finite-temperature static and dynamical proper- ties of spin systems using quantum imaginary time evo- lution, PRX Quantum2, 010317 (2021)

    2021

  20. [28]

    Yeter-Aydeniz, G

    K. Yeter-Aydeniz, G. Siopsis, and R. C. Pooser, Scatter- ing in the Ising model with the quantum Lanczos algo- rithm, New Journal of Physics23, 043033 (2021)

    2021

  21. [29]

    Z. Cai, R. Babbush, S. C. Benjamin, S. Endo, W. J. Hug- gins, Y. Li, J. R. McClean, and T. E. O’Brien, Quantum error mitigation, Reviews of Modern Physics95(2023)

    2023

  22. [30]

    Cotler, S

    J. Cotler, S. Choi, A. Lukin, H. Gharibyan, T. Grover, M. E. Tai, M. Rispoli, R. Schittko, P. M. Preiss, A. M. Kaufman, M. Greiner, H. Pichler, and P. Hayden, Quan- tum virtual cooling, Phys. Rev. X9, 031013 (2019)

    2019

  23. [31]

    W. J. Huggins, S. McArdle, T. E. O’Brien, J. Lee, N. C. Rubin, S. Boixo, K. B. Whaley, R. Babbush, and J. R. McClean, Virtual distillation for quantum error mitiga- tion, Phys. Rev. X11, 041036 (2021)

    2021

  24. [32]

    J. J. Wallman and J. Emerson, Noise tailoring for scalable quantum computation via randomized compiling, Physi- cal Review A94(2016)

    2016

  25. [33]

    D. M. Ceperley, Path integrals in the theory of condensed helium, Rev. Mod. Phys.67, 279 (1995)

    1995

  26. [34]

    W. M. C. Foulkes, L. Mitas, R. J. Needs, and G. Ra- jagopal, Quantum Monte Carlo simulations of solids, Rev. Mod. Phys.73, 33 (2001)

    2001

  27. [35]

    Zhang and H

    S. Zhang and H. Krakauer, Quantum Monte Carlo method using phase-free random walks with Slater de- terminants, Phys. Rev. Lett.90, 136401 (2003)

    2003

  28. [36]

    S. R. White, Minimally entangled typical quantum states at finite temperature, Phys. Rev. Lett.102, 190601 (2009)

    2009

  29. [37]

    E. M. Stoudenmire and S. R. White, Minimally entangled typical thermal state algorithms, New Journal of Physics 12, 055026 (2010)

    2010

  30. [38]

    Paeckel, T

    S. Paeckel, T. K¨ ohler, A. Swoboda, S. R. Manmana, U. Schollw¨ ock, and C. Hubig, Time-evolution methods for matrix-product states, Annals of Physics411, 167998 (2019)

    2019

  31. [39]

    Kormos, M

    M. Kormos, M. Collura, G. Tak´ acs, and P. Calabrese, Real-time confinement following a quantum quench to a non-integrable model, Nature Physics13, 246–249 (2016)

    2016

  32. [40]

    McArdle, T

    S. McArdle, T. Jones, S. Endo, Y. Li, S. C. Benjamin, and X. Yuan, Variational ansatz-based quantum simula- tion of imaginary time evolution, npj Quantum Informa- tion5(2019)

    2019

  33. [41]

    X. Yuan, S. Endo, Q. Zhao, Y. Li, and S. C. Benjamin, Theory of variational quantum simulation, Quantum3, 191 (2019)

    2019

  34. [42]

    A. Ray, E. Swaroop, N. Cao, M. Vasmer, and A. Chowd- hury, Quasiprobabilistic imaginary-time evolution on quantum computers (2025), arXiv:2505.06343 [quant- ph]

    2025 arXiv

  35. [43]

    G. H. Low, T. J. Yoder, and I. L. Chuang, Methodology of resonant equiangular composite quantum gates, Phys. Rev. X6, 041067 (2016)

    2016

  36. [44]

    G. H. Low and I. L. Chuang, Optimal Hamiltonian sim- ulation by quantum signal processing, Phys. Rev. Lett. 118, 010501 (2017)

    2017

  37. [45]

    Arute, K

    F. Arute, K. Arya, R. Babbush, D. Bacon, J. C. Bardin, R. Barends, R. Biswas, S. Boixo, F. G. Brandao, D. A. Buell,et al., Quantum supremacy using a programmable superconducting processor, Nature574, 505 (2019)

    2019

  38. [46]

    Zlokapa, B

    A. Zlokapa, B. Villalonga, S. Boixo, and D. A. Lidar, Boundaries of quantum supremacy via random circuit sampling, npj Quantum Information9, 36 (2023)

    2023

  39. [47]

    Schuster, J

    T. Schuster, J. Haferkamp, and H.-Y. Huang, Random unitaries in extremely low depth (2025), arXiv:2407.07754 [quant-ph]

    2025 arXiv

  40. [48]

    Ma and H.-Y

    F. Ma and H.-Y. Huang, How to construct random uni- taries, inProceedings of the 57th Annual ACM Sympo- sium on Theory of Computing, STOC ’25 (Association for Computing Machinery, New York, NY, USA, 2025) p. 806–809

    2025

  41. [49]

    Campbell, Shorter gate sequences for quantum com- puting by mixing unitaries, Physical Review A95(2017)

    E. Campbell, Shorter gate sequences for quantum com- puting by mixing unitaries, Physical Review A95(2017)

    2017

  42. [50]

    M. B. Hastings, Turning gate synthesis errors into inco- herent errors (2016), arXiv:1612.01011 [quant-ph]

    2016 arXiv

  43. [51]

    Campbell, Random compiler for fast Hamiltonian sim- ulation, Phys

    E. Campbell, Random compiler for fast Hamiltonian sim- ulation, Phys. Rev. Lett.123, 070503 (2019)

    2019

  44. [52]

    A. M. Childs, A. Ostrander, and Y. Su, Faster quantum simulation by randomization, Quantum3, 182 (2019)

    2019

  45. [53]

    C.-H. Cho, D. W. Berry, and M.-H. Hsieh, Doubling the order of approximation via the randomized product for- mula, Physical Review A109, 062431 (2024)

    2024

  46. [54]

    G. H. Low, Halving the cost of quantum multiplexed ro- tations (2021), arXiv:2110.13439 [quant-ph]

    2021 arXiv

  47. [55]

    J. M. Martyn and P. Rall, Halving the cost of quantum algorithms with randomization, npj Quantum Informa- tion11(2025)

    2025

  48. [56]

    Huang, Y

    Y. Huang, Y. Shao, W. Ren, J. Sun, and D. Lv, Efficient quantum imaginary time evolution by drifting real-time evolution: An approach with low gate and measurement complexity, Journal of Chemical Theory and Computa- tion19, 3868–3876 (2023)

    2023

  49. [57]

    Jiang, J

    J. Jiang, J. Leng, and L. Lin, Predicting properties of quantum thermal states from a single trajectory (2026), arXiv:2602.12539 [quant-ph]

    2026

  50. [58]

    C.-F. Chen, M. J. Kastoryano, F. G. S. L. Brand˜ ao, and A. Gily´ en, Quantum thermal state preparation (2023), arXiv:2303.18224 [quant-ph]

    2023 arXiv

  51. [59]

    Duane, A

    S. Duane, A. D. Kennedy, B. J. Pendleton, and D. Roweth, Hybrid Monte Carlo, Physics Letters B195, 216 (1987)

    1987

  52. [60]

    Bou-Rabee and J

    N. Bou-Rabee and J. M. Sanz-Serna, Randomized Hamil- tonian Monte Carlo, The Annals of Applied Probability 27(2017)

    2017

  53. [61]

    Greene, M

    A. Greene, M. Kjaergaard, M. E. Schwartz, G. O. Samach, A. Bengtsson, M. O’Keeffe, D. K. Kim, M. Mar- vian, A. Melville, B. M. Niedzielski, A. Vepsalainen, R. Winik, J. Yoder, D. Rosenberg, S. Lloyd, T. P. Or- lando, I. Marvian, S. Gustavsson, and W. D. Oliver, Error mitigatio...

    2021 arXiv

  54. [62]

    Tao, Structure and randomness in combinatorics (2007), proceedings of the 48th Annual IEEE Sympo- sium on Foundations of Computer Science (FOCS 2007), arXiv:0707.4269 [math.CO]

    T. Tao, Structure and randomness in combinatorics (2007), proceedings of the 48th Annual IEEE Sympo- sium on Foundations of Computer Science (FOCS 2007), arXiv:0707.4269 [math.CO]

    2007 arXiv

  55. [63]

    pseudorandom

    Note that the term “pseudorandom” here refers to its use in additive combinatorics [60], which differs from its meaning in quantum information

  56. [64]

    J. D. Whitfield, J. Biamonte, and A. Aspuru-Guzik, Sim- ulation of electronic structure Hamiltonians using quan- tum computers, Molecular Physics109, 735 (2011)

    2011

  57. [65]

    van den Berg, Z

    E. van den Berg, Z. K. Minev, and K. Temme, Model-free readout-error mitigation for quantum expectation values, 20 Phys. Rev. A105, 032620 (2022)

    2022

  58. [66]

    Y. Zhan, Z. Ding, J. Huhn, J. Gray, J. Preskill, G. K.-L. Chan, and L. Lin, Rapid quantum ground state prepara- tion via dissipative dynamics, Phys. Rev. X16, 011004 (2026)

    2026

  59. [67]

    Martyn and D

    J. Martyn and D. Luo, High-accuracy neural-network quantum states via randomized real-time dynamics, in 39th Conference on Neural Information Processing Sys- tems (NeurIPS 2025) Workshop: Machine Learning and the Physical Sciences Workshop(San Diego, CA, USA, 2025)

    2025

  60. [68]

    Laflamme, J

    R. Laflamme, J. Lin, and T. Mor, Algorithmic cooling for resolving state preparation and measurement errors in quantum computing, Phys. Rev. A106, 012439 (2022)

    2022

  61. [69]

    A. Y. Kitaev, A. Shen, and M. N. Vyalyi,Classical and quantum computation, 47 (American Mathematical Soc., 2002)

    2002

  62. [70]

    Arulandu, J

    A. Arulandu, J. M. Martyn, and I. L. Chuang, Trading Imaginary Time for Randomness in Ground State Prepa- ration [github] (2026)

    2026

  63. [71]

    I. S. Gradshteyn and I. M. Ryzhik,Table of integrals, series, and products(Academic press, 2014)

    2014

  64. [72]

    W. T. Gowers, A new proof of Szemer´ edi’s theorem, Ge- ometric and Functional Analysis11, 465 (2001)

    2001

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.