REVIEW 4 major objections 5 minor 18 references
Simulating Non-Markovian Quantum Dynamics on NISQ Computers Using the Hierarchical Equations of Motion
T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash
Pith's one-line read This paper claims that any non-unitary open-system propagator, in particular the numerically exact HEOM propagator, can be rewritten as a short two-unitary quantum circuit that reproduces exact population dynamics on noisy near-term…
desk verdict A clean, honest NISQ implementation paper: known ingredients combined for HEOM, with real-hardware demos on 2-3 qubits, but no quantum advantage and an unproven scalability claim. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central object is the singular-value decomposition of the non-unitary propagator, written as a linear combination of two unitaries via the diagonal singular-value matrix $\Sigma = (\sigma_0/2)(\Sigma_+ + \Sigma_-)$, where $(\Sigma_\pm)_{jj} = \tilde\sigma_j \pm i\sqrt{1-\tilde\sigma_j^2}$. The circuit then implements $U_\Sigma = \Sigma_+ \oplus \Sigma_-$ as a diagonal unitary using Walsh operators, with Walsh coefficients obtained by a Walsh-Fourier transform, and a single Hadamard-controlled structure yields $G(t)|\Phi(0)\rangle/\sigma_0$ when the ancilla reads 0. The projection operator onto the physical subspace is what reduces the circuit from the full HEOM space to two or three qubits, and this reduction is what makes the hardware results accurate.
What would settle it
Take a model where the HEOM propagator has a singular value exactly or nearly equal to zero, so the construction in eq 4 becomes ill-conditioned, and check whether the circuit output still equals $G(t)|\Phi(0)\rangle/\sigma_0$; a more direct falsifier is to run the 2-qubit circuits at parameter regimes where the exact population approaches zero and show that the measured deviation grows with the largest singular value $\sigma_0$ faster than the sampling-error estimate in eq 50 predicts.
Extended reading notes
Core claim
The central claim is that a non-unitary propagator $G(t)$, here the HEOM propagator in a projected subspace, can be lifted into a unitary circuit through the identity $G(t) = (\sigma_0/2)(U\Sigma_+ V^\dagger + U\Sigma_- V^\dagger)$, where $\Sigma_\pm$ are diagonal unitaries built from the singular values of $G(t)$. The circuit uses one ancilla qubit, and the diagonal unitary $U_\Sigma = \Sigma_+ \oplus \Sigma_-$ is implemented with the Walsh-operator representation, which turns the dominant compilation cost into commuting rotations of tensor products of identity and Pauli-$Z$ operators. Because the projection operator can be chosen freely, the propagator can be restricted to the subspace of the physical quantities of interest, yielding two-qubit circuits for population dynamics. The paper reports that these circuits reproduce numerically exact HEOM population dynamics for the two model systems on both a noiseless circuit simulator and on a real NISQ device, with the largest remaining deviations occurring when the exact population approaches zero.
Load-bearing premise
Everything hinges on the premise that the propagator $G(t)$ is already known classically: the quantum circuit only applies a precomputed matrix to a known vector, so the reported agreement is an implementation check, not an independent computation.
Editorial extensions
If this is right
- The same SVD-plus-Walsh recipe can convert propagators from any quantum master equation into quantum circuits, not only the HEOM propagator.
- Choosing a smaller projection subspace cuts circuit depth and two-qubit gate count enough that NISQ results become nearly exact, as demonstrated by the reduction from 3-qubit to 2-qubit circuits.
- The approach allows independent projection subspaces to be run in parallel, so different physical quantities can be simulated on separate circuits without approximation.
- In the strongly coupled regimes studied, the Lindblad equation is quantitatively unreliable, while qHEOM reproduces the numerically exact dynamics, providing a benchmark for where Markovian master equations break down.
- Circuit depth and two-qubit gate counts are reduced by more than a factor of two relative to Sz.-Nagy dilation for the same propagator.
Reading between the lines
- Because the propagator $G(t)$ is precomputed classically by solving HEOM before the circuit runs, the reported device agreement verifies the circuit implementation rather than demonstrating any quantum speedup; a genuine advantage would require the quantum device to construct or apply the propagator more efficiently than a classical solution.
- The method's practical reach on noisy hardware is tied to finding small projection subspaces, so extending it to large multi-site systems will require either much lower error rates or a systematic way to compress the relevant subspace.
- The paper's construction implies a general compiler: any classically available propagator, from path integrals, tensor-train methods, or generalized quantum master equations, could be substituted for the HEOM propagator without changing the circuit-building procedure.
- A testable extension would be to apply the same circuits to a low-temperature or structured-reservoir model where HEOM requires many effective modes, and compare hardware fidelity against the classical HEOM cost; this would delimit the noise-limited regime of the method.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This paper presents qHEOM, a quantum algorithm for simulating non-Markovian open quantum system dynamics on NISQ devices. The non-unitary propagator G(t) of the hierarchical equations of motion (HEOM) is precomputed classically (eq 33), decomposed via SVD into a sum of two unitaries (eq 5), and implemented with a one-qubit dilation circuit (Fig. 1). The diagonal unitary in the dilation is compiled using a Walsh-operator representation (Section 2.4). The method is demonstrated for a molecular triad charge-transfer model and the FMO complex, using both a noisy simulator (QasmSimulator) and the IBM Sherbrooke device, with dynamic decoupling and twirling error mitigation. The paper also compares HEOM with a Lindblad-type TCL-Redfield equation for the same models, showing that the Lindblad equation fails in these strongly coupled regimes.
Significance. The SVD dilation and subspace projection are mathematically sound and the derivation is clear. The paper is refreshingly honest in stating that the propagator is precomputed classically and that no quantum advantage is claimed; the code is publicly available. If the efficiency of the Walsh compilation held for larger subspaces, the method would be a valuable generic recipe for turning any classically-computable open-system propagator into a short quantum circuit. However, the demonstrations are limited to two- and three-qubit circuits, the agreement with HEOM is forced by construction, and the hardware results lack error bars. The scalability of the Walsh approach for the phase vectors arising from HEOM propagators is not analyzed, so the central claim of a 'generic recipe' for arbitrary master equations on NISQ remains unsubstantiated.
major comments (4)
- [Section 5.3, eq 33] The agreement between qHEOM and HEOM in Figures 7–11 is guaranteed by construction, because the circuit is built from the classically computed HEOM propagator G(t) = M P e^{-iHt} P M†. The statement that 'the excellent agreement ... validates the accuracy of our quantum algorithm' is therefore an overstatement; the figures demonstrate that the circuit implements the desired unitary operations under noise, not that the underlying HEOM dynamics are correct. The paper should either remove or qualify this claim and present the demonstration as an implementation check. This is important because the paper's stated purpose is to show how HEOM can be implemented on quantum circuits, and the validation should be framed accordingly.
- [Section 2.4] The claim that the Walsh operator representation provides an 'efficient' implementation of the diagonal unitary UΣ is not supported for large subspaces. For a generic n-qubit diagonal unitary, the Walsh decomposition requires N = 2^n terms and O(2^n) CNOT gates (Ref. 79). The paper reports circuit depths for n=2 and n=3 only (Tables 3 and 4) and provides no argument that the phase vector f_k of a projected HEOM propagator has a sparse Walsh spectrum. Without such an argument or scaling data, the claim that the method is a 'generic recipe' for arbitrary quantum master equations on NISQ devices is unsubstantiated.
- [Section 5.3, Figures 8–11] The hardware results are presented without error bars, confidence intervals, or a quantitative error metric. With 20,000 shots per time point, the statistical sampling error should be reported to support claims such as 'almost perfectly aligned with numerically exact benchmark results'. Without this information, the reader cannot judge the significance of the observed deviations (or lack thereof).
- [Section 5.3, Table 3] The comparison of circuit complexity with the Sz.-Nagy method (Ref. 63) is based on a single time point and a single model system. The claim that SVD+Walsh reduces circuit complexity by 'more than a factor of 2' is not justified as a general statement; additional time points or model systems are needed to establish the generality of this comparison.
minor comments (5)
- [Eq 49] The extraction formula Pi = σ0 sqrt(Ni/Nc) should be clarified; it implies a specific normalization of the state |Φ(0)>, and the relation between the measured probability and the population should be stated explicitly.
- [Figure 9] The axis label 'Poulation' should be 'Population'.
- [Section 3.3] The notation P|Ψ(0)⟩ = |Ψ(0)⟩ is used as a constraint; it would be helpful to state explicitly that this restricts the initial states to those supported on the subspace S, and that this is a limitation of the method for initial states outside S.
- [Section 2.4] The definition of the Walsh coefficients in eq 12 uses the Hilbert-Schmidt inner product; for readers not familiar with Walsh analysis, the relation to the standard Walsh-Hadamard transform could be stated more explicitly.
- [Section 5.2] The FMO parameters (η = 70 cm^-1, ω_c^-1 = 50 fs) are given in the text but not in a table; a table would improve readability.
Circularity Check
No circularity: qHEOM is a transparent, exact compilation of a classically precomputed HEOM propagator into unitaries, with no fitted parameters and no load-bearing self-citation.
full rationale
The derivation chain is self-contained and non-circular. The paper defines the target by Eq. 1, |Φ(t)⟩ = G(t)|Φ(0)⟩, where G(t) is any non-unitary propagator. For HEOM, G(t) is constructed in Eq. 33, G(t) = MPe^{−iHt}PM†, from the classical HEOM effective Hamiltonian. The SVD step, Eq. 5, G(t) = (σ0/2)(UΣ+V† + UΣ−V†), is an exact matrix identity valid for any propagator, and the Walsh representation, Eqs. 8–13, is an exact expansion of any diagonal unitary. Eq. 7 then shows the circuit output equals G(t)|Φ(0)⟩/σ0 exactly. There is no parameter fitted to any target data; the singular values come directly from the given G(t). The agreement between qHEOM and HEOM in Figures 7–11 is therefore an implementation check of the unitary compilation, not an independent verification of HEOM, but the paper does not disguise this: it states, "It should be noted that the construction of quantum circuits is based on the propagator G(t), which, according to eq 33, is precomputed by solving the HEOM on a classical computer," and "the purpose of this work is not to demonstrate quantum advantage but to illustrate how numerically exact HEOM can be implemented using quantum circuits based on unitary gates." The central mathematical content is externally grounded: HEOM, SVD dilation from Schlimgen et al., and Walsh operator compilation from Welch et al. Self-citations (e.g., Refs. 13, 14, 18, 24, 63, 76) appear for comparisons and methodological continuity, but none is load-bearing for the derivation. The scalability of the Walsh-compiled diagonal unitary is a genuine open question, but it is an efficiency concern, not a circularity. Thus the paper merits a circularity score of 0.
Assumptions & free parameters
free parameters (3)
- HEOM effective mode count K =
K <= 5
- Fock-space truncation L per effective mode =
not specified numerically in text
- Rate-constant fit window =
t = 3000 to 4000 fs
assumptions (6)
- domain assumption HEOM with the Debye spectral density and exponential decomposition of the bath correlation function provides numerically exact dynamics for the spin-boson and FMO models.
- standard math The twin-space thermo-field purification maps the HEOM density matrices to a state vector with a Hermitian effective Hamiltonian (eq 26).
- standard math Singular value decomposition expresses any matrix G as U Σ V† and decomposes Σ into two diagonal unitaries (eqs 2-5).
- standard math Walsh operators form a complete basis for diagonal unitaries, enabling efficient implementation (eqs 8-13).
- domain assumption The projection operator P satisfies P|Ψ(0)⟩=|Ψ(0)⟩ and M†M is identity on the subspace, yielding G(t) = M P e^{-iHt} P M† (eq 33).
- domain assumption Lindblad / TCL-Redfield RWA is the correct Markovian weak-coupling baseline for comparison.
Cite this review
Pith. "Pith review of Simulating Non-Markovian Quantum Dynamics on NISQ Computers Using the Hierarchical Equations of Motion." pith.science (2026). https://pith.science/paper/NTPM4GTQ
@misc{pith2026241112049,
author = {Pith},
title = {Pith review of: Simulating Non-Markovian Quantum Dynamics on NISQ Computers Using the Hierarchical Equations of Motion},
year = {2026},
howpublished = {\url{https://pith.science/paper/NTPM4GTQ}},
note = {Machine review of arXiv:2411.12049}
}
read the original abstract
Quantum computing offers promising new avenues for tackling the long-standing challenge of simulating the quantum dynamics of complex chemical systems, particularly open quantum systems coupled to external baths. However, simulating such non-unitary dynamics on quantum computers is challenging since quantum circuits are specifically designed to carry out unitary transformations. Furthermore, chemical systems are often strongly coupled to the surrounding environment, rendering the dynamics non-Markovian and beyond the scope of Markovian quantum master equations like Lindblad or Redfield. In this work, we introduce a quantum algorithm designed to simulate non-Markovian dynamics of open quantum systems. Our approach enables the implementation of arbitrary quantum master equations on noisy intermediate-scale quantum (NISQ) computers. We illustrate the method as applied in conjunction with the numerically exact hierarchical equations of motion (HEOM) method. The effectiveness of the resulting quantum HEOM algorithm is demonstrated as applied to simulations of the non-Lindbladian electronic energy and charge transfer dynamics in models of the carotenoid-porphyrin-\ce{C60} molecular triad dissolved in tetrahydrofuran and the Fenna-Matthews-Olson complex.
Figures
Figures from the paper (8 more)
Reference graph
Works this paper leans on
-
[19]
(6) Pollard, W. T.; Felts, A. K.; Friesner, R. A. The Redfield equation in condensed-phase quantum dynamics. Adv. Chem. Phys.1996, 93,
work page 1996
-
[75]
(65) Motta, M.; Sun, C.; Tan, A. T.; O’Rourke, M. J.; Ye, E.; Minnich, A. J.; Bran- dao, F. G.; Chan, G. K.-L. Determining eigenstates and thermal states on a quantum computer using quantum imaginary time evolution. Nat. Phys. 2020, 16, 205–210. 47 (66) Kamakari, H.; Sun, S.-N.; Motta, M.; Minnich, A. J. Digital Quantum Simulation of Open Quantum Systems ...
arXiv 2020
-
[183]
(148) Shor, P. W. Scheme for reducing decoherence in quantum computer memory. Phys. Rev. A1995, 52, R2493. (149) Steane, A. M. Active stabilization, quantum computation, and quantum state synthe- sis. Phys. Rev. Lett.1997, 78,
work page 1997
-
[221]
Efficient quantum circuits for diagonal unitaries without ancillas
(79) Welch, J.; Greenbaum, D.; Mostame, S.; Aspuru-Guzik, A. Efficient quantum circuits for diagonal unitaries without ancillas. New. J. Phys.2014, 16, 033040. (80) Walsh, J. L. A closed set of normal orthogonal functions. Am. J. Math. 1923, 45, 5–24. (81) Schipp, F. Walsh series. An introduction to dyadic harmonic analysis; Akad´ emiai Kiad´ o,
work page 2014
-
[265]
(138) Marcus, R. A. Electron Transfer Reactions in Chemistry. Theory and Experiment. Rev. Mod. Phys.1993, 65, 599–610. (139) Javadi-Abhari, A.; Treinish, M.; Krsulich, K.; Wood, C. J.; Lishman, J.; Gacon, J.; 55 Martiel, S.; Nation, P. D.; Bishop, L. S.; Cross, A. W.; Johnson, B. R.; Gambetta, J. M. Quantum computing with Qiskit. arXiv:2405.08810 2024, (1...
arXiv 1993
-
[456]
(126) Moix, J.; Wu, J.; Huo, P.; Coker, D.; Cao, J. Efficient energy transfer in light- harvesting systems, III: The influence of the eighth bacteriochlorophyll on the dy- namics and efficiency in FMO. J. Phys. Chem. Lett.2011, 2, 3045–3052. (127) Sarovar, M.; Cheng, Y. C.; Whaley, K. B. Environmental Correlation Effects on Ex- citation Energy Transfer in...
work page 2011
-
[726]
48 (76) Tong, Z.; Gao, X.; Cheung, M. S.; Dunietz, B. D.; Geva, E.; Sun, X. Charge transfer rate constants for the carotenoid-porphyrin-C60 molecular triad dissolved in tetrahydrofuran: The spin-boson model vs the linearized semiclassical approximation. J. Chem. Phys.2020, 153, 044105. (77) Sweke, R.; Sanz, M.; Sinayskiy, I.; Petruccione, F.; Solano, E. D...
work page 2020
-
[728]
(5) Redfield, A. G. On the theory of relaxation processes. IBM J. Res.1957, 1,
work page 1957
Show all 18 references
-
[948]
Ensemble Method in the Theory of Irreversibility
(108) Zwanzig, R. Ensemble Method in the Theory of Irreversibility. J. Chem. Phys.1960, 33, 1338–1341. (109) Mori, H. Transport, Collective Motion, and Brownian Motion. Prog. Theor. Phys. 1965, 33, 423–455. (110) Koyanagi, S.; Tanimura, Y. Thermodynamic quantum Fokker–Planck e...
1960
-
[1987]
Kondo Memory in Driven Strongly Correlated Quantum Dots
(10) Zheng, X.; Yan, Y.; Di Ventra, M. Kondo Memory in Driven Strongly Correlated Quantum Dots. Phys. Rev. Lett.2013, 111, 086601. (11) Breuer, H.-P.; Laine, E.-M.; Piilo, J.; Vacchini, B. Colloquium: Non-Markovian dy- namics in open quantum systems. Rev. Mod. Phys.2016, 88, 0...
2013
-
[1990]
(82) Ishizaki, A.; Fleming, G. R. Unified Treatment of Quantum Coherent and Incoher- ent Hopping Dynamics in Electronic Energy Transfer: Reduced Hierarchy Equation Approach. J. Chem. Phys.2009, 130, 234111. (83) Lambert, N.; Raheja, T.; Cross, S.; Menczel, P.; Ahmed, S.; Pitch...
2009
-
[2010]
D.; Perdomo-Ortiz, A.; Yung, M.-H.; Aspuru-Guzik, A
(49) Kassal, I.; Whitfield, J. D.; Perdomo-Ortiz, A.; Yung, M.-H.; Aspuru-Guzik, A. Sim- 45 ulating Chemistry Using Quantum Computers. Annu. Rev. Phys. Chem.2011, 62, 185–207. (50) Georgescu, I. M.; Ashhab, S.; Nori, F. Quantum simulation. Rev. Mod. Phys.2014, 86, 153–185. (51...
2011 arXiv
-
[2012]
K.; Bloch, F
(4) Wangsness, R. K.; Bloch, F. The dynamical theory of nuclear induction. Phys. Rev. 1953, 89,
1953
-
[2013]
A short introduction to the Lindblad master equation
(8) Manzano, D. A short introduction to the Lindblad master equation. AIP Adv.2020, 10, 025106. (9) Alicki, R.; Lendi, K. Quantum Dynamical Semigroups and Applications; Springer- Verlag: Berlin,
2020
-
[2252]
J.; Munro, W
56 (150) Devitt, S. J.; Munro, W. J.; Nemoto, K. Quantum error correction for beginners. Rep. Prog. Phys.2013, 76, 076001. (151) Gottesman, D. An introduction to quantum error correction and fault-tolerant quan- tum computation. Quantum information science and its contribution...
2013
-
[3301]
W.; Head-Marsden, K.; Sager, L
(58) Schlimgen, A. W.; Head-Marsden, K.; Sager, L. M.; Narang, P.; Mazziotti, D. A. 46 Quantum Simulation of Open Quantum Systems Using a Unitary Decomposition of Operators. Phys. Rev. Lett.2021, 127, 270503. (59) Schlimgen, A. W.; Head-Marsden, K.; Sager, L. M.; Narang, P.; M...
2021
-
[3322]
R.; Pollock, F
(153) Jørgensen, M. R.; Pollock, F. A. Exploiting the Causal Tensor Network Structure of Quantum Processes to Efficiently Simulate Non-Markovian Path Integrals. Phys. Rev. Lett. 2019, 123, 240602. (154) Richter, M.; Hughes, S. Enhanced TEMPO Algorithm for Quantum Path Integral...
2019 arXiv
-
[9591]
D.; Geva, E.; Sun, X
(120) Brian, D.; Liu, Z.; Dunietz, B. D.; Geva, E.; Sun, X. Three-state harmonic models for photoinduced charge transfer. J. Chem. Phys.2021, 154, 174105. (121) Brian, D.; Sun, X. Charge-Transfer Landscape Manifesting the Structure–Rate Re- lationship in the Condensed Phase Vi...
2021
Reviewed August 12, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.