Pith.

Quantum Physics · arXiv:2611.17036 v1 · submitted 12 Nov 2026 · 51 references

Ergodicity breaking for data classification in a dissipative Ising model with competing two- and four-body interactions

Simon Kochsiek, Mario Boneberg, Igor Lesanovsky

The paper says

“We explore the close relationship between the processing of data and the dynamics of interacting many-body systems.”

more of the abstract

For example, certain classes of quantum neural networks can be mapped onto dissipative quantum Ising models whose evolution is described by a Lindblad master equation. In these architectures, emergent nonlinearities lead to non-ergodic dynamics under which initial states located in different basins of attraction evolve towards distinguishable stationary states. This mechanism underlies the classification of quantum data. Motivated by this connection, we consider an Ising model with competing two- and four-body interactions. Using a combination of mean-field calculations and exact small-scale numerics, we show that the stationary state manifold can be designed to enable multi-class classification. We also investigate the structure of the ensuing basins of attraction, thereby showing that these systems allow for the classification of quantum states with nonlinear decision boundaries.

from the abstract

Pith says

Machine review · deepseek-v4-flash · 13 Nov 2026 · what was read was not recorded · not checked by a person

It can be wrong. Challenge this review or reply to it.

Holds up, with conditions. Mean-field analysis of a dissipative Ising model with four-body interactions shows up to five stable fixed points, but the multi-class classification claim rests entirely on the mean-field ansatz and is not demonstrated in the exact dynamics or on an actual classification task. This line is wrong

3 major objections · 7 minor

How often objections from reviews like this one are right: not yet measured

Are you an author? What you can do.

Not checked: figures, code, what the 51 references say.

The paper says, at Sec. IV, Eq. (12) and Sec. V

The review did not quote this passage.

A classifier read this sentence of the paper as acknowledging the concern (probability 0.83):

“A limiting aspect of our investigation is certainly that the majority of its findings are based on a mean-field analysis.”

Pith says: Objection 1 of 3 · major

The central claim of multi-class classification (up to 5 stable stationary states) rests entirely on the mean-field analysis with the product-state ansatz (Eq. 12). The finite-size N=8 numerics (Sec. V) cannot break ergodicity, so the authors can only report qualitative signatures in magnetization distributions, not a demonstration of multi-class classification.

more of objection 1

The authors acknowledge this limitation in Sec. VI, but the paper does not provide any evidence that the mean-field fixed-point structure translates to the exact Liouvillian in the thermodynamic limit. Without such evidence, the claim that the model 'enables multi-class classification' is unsupported beyond mean-field. A concrete classification example (e.g., assigning input states to classes based on stationary states) is also missing.

The paper says, at Sec. IV B 1, Eqs. (23)-(24)

The review did not quote this passage.

Pith says: Objection 2 of 3 · major

The derivation of the cubic equations for uniform and staggered fixed points assumes the product-state ansatz and the two-sublattice structure. While internally consistent, the paper does not discuss whether other types of fixed points (e.g., with more complex sublattice structures or non-product states) could exist and affect the classification claim. The mean-field closure is a strong assumption, and the paper would benefit from a discussion of its validity beyond the product-state ansatz.

The paper says, at Sec. V, Fig. 5

The review did not quote this passage.

A classifier read this sentence of the paper as acknowledging the concern (probability 0.75):

“Small-scale numerical calculations provided qualitative evidence for the applicability of the conclu- sions drawn from mean-field.”

Pith says: Objection 3 of 3 · major

The finite-size data (N=8) show that ⟨m_x⟩ always decays to zero, and the classification claim is inferred from the shape of the distribution of m_x (second moment and relative weights). However, the paper does not demonstrate that these signatures correspond to actual classification of initial states into distinct classes. The mapping from mean-field basins to finite-size behavior is qualitative at best. The authors call for tensor-network studies, but the current evidence is insufficient to support the multi-class classification claim.

7 minor objections

The paper says, at General

Pith says: minor objection 1 of 7

The paper is well-written and the figures are clear. The code and data are openly available, which is commendable.

The paper says, at Sec. II, Eq. (4)

Pith says: minor objection 2 of 7

The rescaling sets W=1, but the paper later uses W=1 in all figures. This is fine, but it would be helpful to state explicitly that W is positive and used as the energy unit, and that the results are for W>0.

show 5 more

The paper says, at Sec. III A, Eq. (5)

Pith says: minor objection 3 of 7

The parity operator P is defined as product of σ^z_k. It might be helpful to note that this is a global Z2 symmetry.

The paper says, at Sec. IV B 1, Eq. (23)

Pith says: minor objection 4 of 7

The cubic polynomial for u = m_x^2 is given. It would be helpful to explicitly state that u ≥ 0 is required for physical solutions.

The paper says, at Sec. IV C 2, Fig. 4

Pith says: minor objection 5 of 7

The basin plots are informative, but the sampling is restricted to initial states with y=0 and pure states. The paper could note that this restriction may miss some basin structure.

The paper says, at Sec. V, Eq. (32)

Pith says: minor objection 6 of 7

The relative weight w(μ_x) is defined. It might be helpful to note that w=1 corresponds to the paramagnetic distribution.

The paper says, at Appendix C

Pith says: minor objection 7 of 7

The discussion of the emergent symmetry and critical manifold is interesting but somewhat dense. A brief summary in the main text would help readability.

What would settle it

Perform a tensor-network simulation (e.g., using infinite MPS or PEPS) for the dissipative Ising model with two- and four-body interactions at the parameter point (Ω,V) = (-0.05,0.05) (point ◇ in Fig. 2(b)), where mean-field predicts five stable fixed points. Compute the steady-state expectation values ⟨m_x⟩ and ⟨n_x⟩ for a large system (e.g., N=64 or infinite) and check whether multiple distinct stationary states emerge, or whether the system relaxes to a unique steady state. If multiple stationary states are found, verify that their magnetization values match the mean-field predictions. If only a unique steady state is found, the multi-class classification claim is unsupported.

from the skeptic's reading

How the review went

4 readings by deepseek-v4-flash · 13 Nov 2026

  1. Reader holds up, with conditions confidence moderate
  2. Skeptic agrees 10 s · 22,235 tokens read, 718 written
  3. Editor wrote the verdict above 14 s · 23,171 tokens read, 596 written
  4. Referee major revision: 3 major objections, 7 minor 14 s · 22,479 tokens read, 1,151 written

Not run on this paper: the simulated authors' defence.

Pith did not check this review's objections against the paper.

The review in full

The review in brief, machine-written

Simon,

This paper adds a four-body interaction term to the dissipative Ising model studied in Refs. 33 and 42 and works out the mean-field fixed-point structure. The new result is the root-count analysis of the cubics (23)-(24), which yields up to nine fixed points, five of them stable, and the basin decomposition of the two-sublattice mean-field state space into up to five attractors. The derivation is internally consistent—I re-derived the fixed-point equations and the stability boundary (27) and they match. The authors are also unusually honest about the limitations: Sec. VI states plainly that most findings rest on mean-field theory and that tensor-network studies are needed, and Sec. V confines itself to qualitative 'signatures' in N=8 exact numerics.

The soft spot is the one the authors themselves flag: the multi-class classification claim is a property of the six mean-field equations (14)-(16), not of the exact Liouvillian (1). The two-sublattice product-state ansatz (12) is assumed for all times, and the finite-size data (N=8) cannot break ergodicity, so the paper can only report magnetization-distribution 'signatures' that are consistent with the mean-field picture but do not demonstrate that the multi-stability survives in the thermodynamic limit. No actual classification task is executed—the abstract and title claim a classification capability, but the evidence is the mean-field basin structure plus qualitative N=8 histograms. The overlap with Ref. 43 on multi-site interactions could not be checked from the text, but that is a minor concern given the paper's own derivations.

Who this is for: researchers working on dissipative quantum neural networks and mean-field phase diagrams of open spin systems. The mean-field analysis is solid and the phase diagram is a useful reference for that community. The paper deserves a serious referee because the derivations are correct and the question—whether multi-body interactions can increase the number of metastable states—is well-motivated. The referee should ask for a concrete classification demonstration or larger-scale numerics before the classification claim can be accepted, but the mean-field results themselves are worth publishing.

Recommendation: send to peer review. The mean-field work is sound, the limitations are stated, and the paper is a reasonable step forward even if the central claim is conditional on future work.

The strongest objection

The central claim—that adding a four-body term enables multi-class classification with up to five stable stationary states—rests entirely on the mean-field analysis of Sec. IV. The two-sublattice product-state ansatz (Eq. 12) is assumed for all times, and the fixed-point counts, stability regions, and basin structures are properties of the six mean-field equations (14)-(16), not of the exact Liouvillian (1). The authors acknowledge this limitation explicitly in Sec. VI ("A limiting aspect of our investigation is certainly that the majority of its findings are based on a mean-field analysis") and in Sec. V, where they note that for N=8 ergodicity cannot be broken and they can only report qualitative magnetization-distribution 'signatures'. The finite-size data (Fig. 5) show that the stationary state always has ⟨m_x⟩=0, and the claimed multi-class capability is inferred from the shape of the distribution of m_x, not from any actual classification task. The paper does not demonstrate that the mean-field multi-stability translates into a genuine thermodynamic-limit property of the Liouvillian, nor does it provide a concrete classification example. The authors themselves call for tensor-network studies to settle this. Therefore, the load-bearing concern is that the multi-class classification claim is supported only by mean-field theory, with no evidence that the exact dynamics in the thermodynamic limit realizes the same fixed-point structure.

The reader's weakest_assumption correctly identifies that the multi-stationary-state claims are properties of the mean-field equations, not of the exact Liouvillian. The paper's own limitations (Sec. VI) and finite-size results (Sec. V) confirm that the evidence for multi-class classification is indirect. The reader's verdict of CONDITIONAL is appropriate because the mean-field derivations are internally consistent, but the classification claim requires additional support. My stress-test does not identify a new concern beyond what the reader already noted; rather, it reinforces the same point with more concrete detail about the gap between mean-field predictions and exact dynamics. The agreement is 'agree' because the weakest_assumption is the same. The verdict should remain CONDITIONAL, as the paper's own suggested future work (tensor-network studies) would be the natural path to resolve the concern.

The referee's report

If the mean-field multi-stability survives in the thermodynamic limit, the work provides a concrete mechanism for multi-class classification in dissipative quantum neural networks, extending previous binary classification results. The explicit demonstration of up to 5 stable fixed points and nonlinear basin boundaries is a clear step forward. The paper is honest about limitations and provides reproducible code/data.

Pith's reading of the paper (a machine's summary, not the authors')

The paper studies a chain of tiny magnets (quantum spins) that lose energy by decaying, are pushed by a drive, and interact with their neighbors in pairs and in groups of four. In a single chain the dynamics eventually settles into one resting state. But the paper looks at an approximate description of a very long chain, called mean-field theory, where the competition between driving, decay, and the two- and four-spin couplings can produce several resting configurations at once. Up to nine configurations exist in this approximation, and up to five of them are stable.

Every starting state gets pulled toward one of these stable configurations. Which one it reaches divides the space of starting states into regions, called basins of attraction. The boundaries between basins are curved and complicated, which in machine learning language means the system can implement nonlinear decision boundaries. With two-spin couplings alone there are at most two attracting states, which supports binary classification. The four-spin coupling adds extra stable states, so the same mechanism could sort inputs into four or five classes.

The honest caveat: these multiple resting states only show up in the mean-field approximation. In a small exact simulation with eight spins, the system always reaches one unique state, and the paper can only see indirect hints of the multiple states in how the magnetization is distributed. The authors state this limitation explicitly and propose tensor-network calculations as the next step. So the paper offers a consistent mechanism and a clear target for what larger calculations should test, not yet proof that the phenomenon survives in the full quantum dynamics.

Core claim. Sec. VI states: 'We showed that their incorporation indeed enables multi-class classification manifesting in the presence of up to five stable stationary states for the specific model at hand.' Together with Sec. IV C.2, which shows initial states relax into one of up to five stable fixed points whose basin boundaries are 'highly nonlinear'. If correct, adding a four-body term to the W=0 dissipative Ising model grows the fixed-point manifold from at most 3 (binary classification, Ref. 33) to up to 9 total, 5 stable, enabling a four- or five-class partition of the input state space.

The premise it rests on. The two-sublattice product-state ansatz of Sec. IV A (Eq. 12), rho = rho_A ⊗ rho_B ⊗ rho_A ⊗ ..., assumed for all times, is the only route to every multi-stationary-state claim. The up-to-9/up-to-5 fixed-point counts and the nonlinear basin structure are properties of the six mean-field equations (14)-(16), not of the exact Liouvillian (1). The finite check (N=8, Sec. V) cannot break ergodicity, so the authors can only report qualitative magnetization-distribution 'signatures'. If thermodynamic-limit dynamics do not realize the mean-field multi-stability, the multi-class classification claim collapses. The authors assign this check to future tensor-network studies (Sec. VI).

Check this review

Pith stored no record of this review's passes and took no fingerprint of it. Reviews made by code from 13 Nov 2026 on carry both.

How often reviews like this one are right

Reviews like this one: written by deepseek-v4-flash, where what was read was not recorded.

How often objections from reviews like this one are right: not yet measured

Every value above, in the review's public record (JSON)

Reviews by people

This review was written by a machine. No person has reviewed this paper on Pith yet. A scientist listed on Pith can review it for a fee, and their review is published here whatever it concludes.

Request a human review

How Pith checks papers · The full audit of this paper

The paper

Abstract

We explore the close relationship between the processing of data and the dynamics of interacting many-body systems. For example, certain classes of quantum neural networks can be mapped onto dissipative quantum Ising models whose evolution is described by a Lindblad master equation. In these architectures, emergent nonlinearities lead to non-ergodic dynamics under which initial states located in different basins of attraction evolve towards distinguishable stationary states. This mechanism underlies the classification of quantum data. Motivated by this connection, we consider an Ising model with competing two- and four-body interactions. Using a combination of mean-field calculations and exact small-scale numerics, we show that the stationary state manifold can be designed to enable multi-class classification. We also investigate the structure of the ensuing basins of attraction, thereby showing that these systems allow for the classification of quantum states with nonlinear decision boundaries.

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. No sentence in this record was written by a model.

abstract compiled · 2026-11-13

The authors' main claim, as the classifier picked it out 0.86

“Using a combination of mean-field calculations and exact small-scale numerics, we show that the stationary state manifold can be designed to enable multi-class classification.”

How the main claim is stated: asserted 1.00

What the abstract itself states, as probabilities: claims first 0.03 · states limitation 0.07 · human participants 0.02 · controlled comparison 0.02 · randomized assignment 0.01 · states quantitative result 0.05 · compares with prior results 0.06 · mentions code or data availability 0.02

The abstract, sentence by sentence

This table labels the abstract only (arXiv listing abstract v1). The number is the probability the classifier gave the label.

SentenceRoleThe authors' own workStates the main conclusionStrength
We explore the close relationship between the processing of data and the dynamics of interacting many-body systems. background 0.97 0.33 0.04 asserted 0.96
For example, certain classes of quantum neural networks can be mapped onto dissipative quantum Ising models whose evolution is described by a Lindblad master equation. background 1.00 0.07 0.03 asserted 0.66
In these architectures, emergent nonlinearities lead to non-ergodic dynamics under which initial states located in different basins of attraction evolve towards distinguishable stationary states. background 0.91 0.11 0.11 asserted 1.00
This mechanism underlies the classification of quantum data. interpretation 0.56 0.09 0.09 asserted 1.00
Motivated by this connection, we consider an Ising model with competing two- and four-body interactions. contribution 0.49 0.45 0.03 asserted 0.75
Using a combination of mean-field calculations and exact small-scale numerics, we show that the stationary state manifold can be designed to enable multi-class classification. result 0.57 0.97 0.57 asserted 1.00
We also investigate the structure of the ensuing basins of attraction, thereby showing that these systems allow for the classification of quantum states with nonlinear decision boundaries. result 0.89 0.95 0.39 asserted 1.00

What kind of paper the classifier reports this is: theoretical result 0.88 · evidence from simulation or computation 0.93

What was examined, and what was not

Sources read

  • arXiv listing abstract v1

Not examined

  • full text
  • figures and tables
  • references and citing sentences
  • relations to other papers

Levels run

  • P abstract

Not run

  • S sections
  • T body sentences
  • C citations
  • R cross-paper relations

How the labels were made

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-11-13: 1 request, 5,504 input tokens.

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. This record as JSON

References (51)
  1. [1] C. M. Bishop,Pattern recognition and machine learning (Springer, New York, 2006) on Pith
  2. [2] K. P. Murphy,Machine Learning: A Probabilistic Per- spective(MIT Press, Cambridge, MA, 2012) on Pith
  3. [3] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, Proc. IEEE86, 2278 (1998) on Pith
  4. [4] M. A. Nielsen,Neural Networks and Deep Learning(De- termination Press, 2015) on Pith
  5. [5] A. Rajkomar, J. Dean, and I. Kohane, N. Engl. J. Med. 380, 1347 (2019) on Pith
  6. [6] G. Gunˇ car, M. Kukar, M. Notar, M. Brvar, P. ˇCernelˇ c, et al., Sci. Rep.8, 411 (2018) on Pith
  7. [7] W. Ning, Z. Wang, Y. Gu, L. Huang, S. Liu, Q. Chen, Y. Yang, and G. Hong, Sci. Rep.15, 27857 (2025) on Pith
  8. [8] Y. LeCun, Y. Bengio, and G. Hinton, Nature521, 436 (2015) on Pith
  9. [9] I. Goodfellow, Y. Bengio, and A. Courville,Deep Learn- ing(MIT Press, Cambridge, MA, 2016) on Pith
  10. [10] K. Hornik, M. Stinchcombe, and H. White, Neural Netw. 2, 359 (1989) on Pith
  11. [11] J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, and S. Lloyd, Nature549, 195 (2017) on Pith
  12. [12] V. Dunjko and H. J. Briegel, Rep. Prog. Phys.81, 074001 (2018) on Pith
  13. [13] M. Cerezo, A. Arrasmith, R. Babbush, S. C. Benjamin, S. Endo, K. Fujii, J. R. McClean, K. Mitarai, X. Yuan, L. Cincio, and P. J. Coles, Nat. Rev. Phys.3, 625 (2021) on Pith
  14. [14] M. Cerezo, G. Verdon, H.-Y. Huang, L. Cincio, and P. J. Coles, Nat. Comput. Sci.2, 567 (2022) on Pith
  15. [15] Y. Wang and J. Liu, Rep. Prog. Phys.87, 116402 (2024) on Pith
  16. [16] M. Larocca, S. Thanasilp, S. Wang, K. Sharma, J. Bia- monte, P. J. Coles, L. Cincio, J. R. McClean, Z. Holmes, and M. Cerezo, Nat. Rev. Phys.7, 174 (2025) on Pith
  17. [17] V. Havl ´ ıˇ cek, A. D. C´ orcoles, K. Temme, A. W. Harrow, A. Kandala, J. M. Chow, and J. M. Gambetta, Nature 567, 209 (2019) on Pith
  18. [18] J. Herrmann, S. Masot Llima, A. Remm, P. Zapletal, N. A. McMahon, C. Scarato, F. Swiadek, C. K. Andersen, C. Hellings, S. Krinner,et al., Nat. Commun.13, 4144 (2022) on Pith
  19. [19] H.-Y. Huang, M. Broughton, J. Cotler, S. Chen, J. Li, M. Mohseni, H. Neven, R. Babbush, R. Kueng, J. Preskill, and J. R. McClean, Science376, 1182 (2022) on Pith
  20. [20] I. M. Georgescu, S. Ashhab, and F. Nori, Rev. Mod. Phys.86, 153 (2014) on Pith
show 31 more
  1. [21] J. Preskill, Quantum2, 79 (2018) on Pith
  2. [22] R. O’Donnell and J. Wright, inProceedings of the 48th Annual ACM Symposium on Theory of Computing (STOC)(2016) p. 899 on Pith
  3. [23] J. Haah, A. W. Harrow, Z. Ji, X. Wu, and N. Yu, IEEE Trans. Inf. Theory63, 5628 (2017) on Pith
  4. [24] I. Cong, S. Choi, and M. D. Lukin, Nat. Phys.15, 1273 (2019) on Pith
  5. [25] K. Beer, D. Bondarenko, T. Farrelly, T. J. Osborne, R. Salzmann, D. Scheiermann, and R. Wolf, Nat. Com- mun.11, 808 (2020) on Pith
  6. [26] A. Abbas, D. Sutter, C. Zoufal, A. Lucchi, A. Figalli, and S. Woerner, Nat. Comput. Sci.1, 403 (2021) on Pith
  7. [27] J. J. Meyer, M. Mularski, E. Gil-Fuster, A. A. Mele, F. Arzani, A. Wilms, and J. Eisert, PRX Quantum4, 010328 (2023) on Pith
  8. [28] M. Schuld, I. Sinayskiy, and F. Petruccione, Quantum Inf. Process.13, 2567 (2014) on Pith
  9. [29] Y. Cao, G. G. Guerreschi, and A. Aspuru-Guzik, Quantum neuron: an elementary building block for machine learning on quantum computers (2017), arXiv:1711.11240 [quant-ph] on PithPith's reviewarXiv
  10. [30] Z. Holmes, N. J. Coble, A. T. Sornborger, and Y. Suba¸ sı, Phys. Rev. Res.5, 013105 (2023), arXiv:2112.12307 [quant-ph] on PithPith's reviewarXiv
  11. [31] P. Argyle, D. Lakhdar-Hamina, S. H. Miller, and V. Gal- itski, Measurement-induced quantum neural network (2026), arXiv:2603.19200 [quant-ph] on PitharXiv
  12. [32] M. Boneberg, F. Carollo, and I. Lesanovsky, Phys. Rev. A111, 062405 (2025) on Pith
  13. [33] M. Boneberg, S. Kochsiek, G. Perfetto, and I. Lesanovsky, Phys. Rev. A114, 012209 (2026) on Pith
  14. [34] M. Marcuzzi, E. Levi, S. Diehl, J. P. Garrahan, and I. Lesanovsky, Phys. Rev. Lett.113, 210401 (2014) on Pith
  15. [35] K. Macieszczak, M. Gut ¸˘ a, I. Lesanovsky, and J. P. Gar- rahan, Phys. Rev. Lett.116, 240404 (2016) on Pith
  16. [36] F. Minganti, A. Biella, N. Bartolo, and C. Ciuti, Phys. Rev. A98, 042118 (2018) on Pith
  17. [37] J. J. Hopfield, Proceedings of the National Academy of Sciences79, 2554 (1982) on Pith
  18. [38] D. J. Amit, H. Gutfreund, and H. Sompolinsky, Phys. Rev. A32, 1007 (1985) on Pith
  19. [39] D. J. Amit, H. Gutfreund, and H. Sompolinsky, Phys. Rev. Lett.55, 1530 (1985) on Pith
  20. [40] D. J. Amit, H. Gutfreund, and H. Sompolinsky, Ann. Phys.173, 30 (1987) on Pith
  21. [41] H. Weimer, Phys. Rev. Lett.114, 040402 (2015) on Pith
  22. [42] V. R. Overbeck, M. F. Maghrebi, A. V. Gorshkov, and H. Weimer, Phys. Rev. A95, 042133 (2017) on Pith
  23. [43] P. Wang and R. Fazio, Phys. Rev. A103, 013306 (2021) on Pith
  24. [44] B. Buˇ ca and T. Prosen, New J. Phys.14, 073007 (2012) on Pith
  25. [45] V. V. Albert and L. Jiang, Phys. Rev. A89, 022118 (2014) on Pith
  26. [46] S. Lieu, R. Belyansky, J. T. Young, R. Lundgren, V. V. Albert, and A. V. Gorshkov, Phys. Rev. Lett.125, 240405 (2020) on Pith
  27. [47] M. F. Maghrebi and A. V. Gorshkov, Phys. Rev. B93, 014307 (2016) on Pith
  28. [48] N. Goldenfeld,Lectures on phase transitions and the renormalization group(CRC Press, 2018) on Pith
  29. [49] E. Gillman, F. Carollo, and I. Lesanovsky, Phys. Rev. E 107, L022102 (2023) on Pith
  30. [50] M. Boneberg, F. Carollo, and I. Lesanovsky, New J. Phys. 25, 093020 (2023) on Pith
  31. [51] S. Kochsiek, M. Boneberg, and I. Lesanovsky, Data and code for “Ergodicity breaking for data classification in a dissipative Ising model with competing two- and four- body interactions”, Zenodo,https://doi.org/10.5281/ zenodo.24164117(2026). 12 Appendix A: Mean-field equations... on Pith

Challenges and discussion

Disagree with the review, or know something it missed? Leave a comment, or challenge this review formally: a challenge becomes a permanent entry in this paper's record, and its decision is published here.

No comments yet.

Sign in to comment