Pith. sign in

REVIEW 3 major objections 4 minor 71 references

Many-body dynamical localization in the kicked Bose-Hubbard chain

T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Pith's one-line read A clean, periodically kicked Bose-Hubbard chain can host a many-body dynamically localized phase that persists as the system grows: Floquet states violate eigenstate thermalization and energy absorption stays below infinite temperature.

desk verdict A coherent numerical case for many-body dynamical localization in a clean kicked Bose-Hubbard chain; the thermodynamic-limit claim is plausible but rests on sizes too small to rule out a prethermal plateau. read the letter →

arxiv 1908.03399 v2 pith:VYMLYMBR submitted 2019-08-09 cond-mat.quant-gas cond-mat.stat-mechquant-ph

classification cond-mat.quant-gascond-mat.stat-mechquant-ph
keywords many-bodydynamicallocalizationkickedBose-HubbardmodelFloquetstateseigenstatethermalizationhypothesisentanglemententropyergodicitybreakingperiodicdrivinginfinitetemperature
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 tries to establish that a clean, periodically kicked Bose-Hubbard model — with no disorder — can exhibit many-body dynamical localization: after a brief transient the system stops absorbing energy and its local observables never reach the infinite-temperature value, for almost all initial states. The authors argue this is a genuine phase that survives in the thermodynamic limit, involving every Floquet state, and that it is different from many-body localization because the half-chain entanglement entropy grows linearly in time instead of logarithmically. If the claim is right, it would be a rare example of a non-integrable driven many-body system that evades the 'heating to infinite temperature' fate through quantum interference alone, with the Floquet states delocalized in space yet confined to narrow quasi-degenerate subspaces of the undriven interaction energy.

What carries the argument

The central object is the Floquet-state structure of the unitary step $\hat U = e^{-i\hat V \tau} e^{-i\hat K}$, where $\hat V = \frac{U}{2}\sum_j \hat n_j(\hat n_j - 1)$ is the on-site interaction (diagonal in the Fock basis) and $\hat K = J \sum_j (\hat a^\dagger_j \hat a_{j+1} + \mathrm{h.c.})$ is the hopping delivered by an instantaneous kick. The argument is carried by three diagnostics built from these Floquet states: the heating parameter $\epsilon$ comparing the infinite-time stroboscopic average of $\hat V$ with its $T=\infty$ value; the broadness measure $\delta$ of the distribution of $V_\alpha = \langle \varphi_\alpha | \hat V | \varphi_\alpha \rangle$, which distinguishes ETH from its violation; and the comparison between the inverse participation ratio and its minimal 'delocalized' value $\mathrm{IPR}^{\mathrm{deloc}}$, which shows that states fill the degenerate $V$-subspaces they sit in. The half-chain entanglement entropy $S_{L/2}$ completes the picture by separating this phase from MBL through its linear-in-time growth.

What would settle it

Compute $\delta$ and $\epsilon$ for $L \ge 14$ with the same small-$J$ parameters using an approximation whose truncation error is controlled; if $\delta$ starts bending downward toward zero or $\epsilon$ creeps upward toward one as $L$ grows, the apparent MBDL is a finite-size or prethermal effect. Likewise, if the revival time stops growing linearly with $L$ or the entanglement growth crosses over from linear to logarithmic at larger sizes, the claimed phase is falsified.

Watch

Extended reading notes

Core claim

For hopping strength $J \lesssim 0.05$ (with $U=1$, period $\tau=1$, unit filling), the stroboscopic dynamics of the kicked Bose-Hubbard chain shows ergodicity breaking that persists up to the largest sizes considered: the infinite-time average of the interaction energy $\bar V$ stays well below its $T=\infty$ value, so the normalized heating parameter $\epsilon$ converges to a number smaller than one as $L$ grows. The Floquet states violate eigenstate thermalization in a size-robust way: the distribution of their $V_\alpha$ values stays broad, the diagnostic $\delta = |\langle V\rangle - \exp\langle \log V\rangle|$ bends upward rather than decaying to zero for $J \lesssim 0.05$, and the same behaviour is found starting from an excited initial state, indicating the whole spectrum is affected. Half-chain entanglement entropy grows linearly in time and saturates to a volume-law value much smaller than the thermal one — the opposite of the logarithmic growth seen when disorder is added to induce many-body localization. Floquet states in this phase are delocalized within quasi-degenerate eigenspaces of $V$ (their inverse participation ratio scales with the 'delocalized' benchmark $\mathrm{IPR}^{\mathrm{deloc}}$ raised to a positive power), but they are localized across those subspaces; the authors interpret this as spatial delocalization coexisting with energy localization, which blocks full exploration of the Hilbert space.

Load-bearing premise

The load-bearing premise is that the behaviour seen for $J \lesssim 0.05$ at sizes up to about nineteen sites is the thermodynamic-limit behaviour: that the upward bend of $\delta$ and the saturation of $\epsilon$ below one are signs of a true localized phase, not a prethermal plateau that would give way to heating at larger sizes or later times.

Editorial extensions

If this is right

  • Periodically driven clean lattice systems can be protected from heating to infinite temperature without disorder, so driving alone does not force a trivial infinite-temperature state; the transition at $J \approx 0.06$–$0.1$ is an eigenstate phase transition in the Floquet spectrum.
  • Because the entanglement-entropy growth is linear rather than logarithmic, an experiment that measures entanglement can directly distinguish many-body dynamical localization from many-body localization in the same platform.
  • The revival time of the initial-state overlap grows linearly with $L$, giving an observable finite-size echo signature that is absent in the thermalizing regime.
  • The truncated-Wigner comparison predicts that any classical or decohered version of the same dynamics heats diffusively, so the localization should be destroyed by loss of quantum coherence — a testable prediction for experiments.

Reading between the lines

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

  • If the phase survives at larger sizes, it points to an emergent integrable structure: an extensive set of conserved or nearly conserved quantities in the driven system, whose generators are localized in energy rather than in space.
  • The same mechanism may transfer to other driven lattice models whose undriven part has highly degenerate eigenspaces and whose kick is a hopping operator; a direct test would be to apply the same $\epsilon$ and $\delta$ scalings to a kicked Fermi-Hubbard or spin-1/2 chain.
  • A sharper test of the phase would be to measure correlation spreading: if the picture of delocalization inside degenerate subspaces is right, single-particle correlations should propagate ballistically while energy correlations remain frozen, giving a two-speed light cone.
  • The finite-size data leave open the possibility of a prethermal plateau for $J$ just below the crossover; pushing the same observables to $L \gtrsim 20$ with tensor networks, looking for a downturn of $\delta$ or an upturn of $\epsilon$, would settle whether the phase extrapolates.
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 / 4 minor

Summary. The paper studies a periodically kicked Bose-Hubbard chain with on-site interaction V and kicked hopping K. Using exact diagonalization for L≤11, a ΔT-truncated Hilbert-space method and Krylov propagation for larger sizes, and tDMRG for the overlap and entanglement dynamics, the authors present evidence for a many-body dynamically localized (MBDL) phase at small hopping J≲0.05. In this regime, the infinite-time average of the interaction energy remains below the infinite-temperature value, the distribution of Floquet-state expectation values of V stays broad (δ does not decay with L), the overlap with the initial state remains O(1) with revivals at times linear in L, and the half-chain entanglement entropy grows linearly in time before saturating at a subthermal volume-law value. The authors argue that all Floquet states violate eigenstate thermalization, that the phase persists in the thermodynamic limit, and that it is distinct from MBL because of the linear entanglement growth and the delocalized structure of Floquet states within quasi-degenerate eigenspaces of V. A truncated-Wigner calculation shows that the classical counterpart always diffuses, supporting the quantum nature of the effect.

Significance. If the central claim holds, this would be one of the few disorder-free, translation-invariant driven systems exhibiting ergodicity breaking that survives the thermodynamic limit, and the linear-in-time entanglement growth would sharply distinguish this phase from MBL. The paper is carefully executed: it uses multiple complementary diagnostics (ϵ, δ, overlap, entanglement entropy, IPR), reports honest caveats about finite sizes, checks the Hilbert-space truncation with a border projector, and provides error estimates. The main significance is therefore high, provided the thermodynamic-limit extrapolation can be made more robust.

major comments (3)
  1. [Sec. III A, Fig. 3] The upward bend of δ for J≲0.05, interpreted as evidence for MBDL, is observed only up to L≈11 in the full exact-diagonalization data, and the larger-L points in Fig. 1 rely on the Hilbert-space truncation of Sec. II B, whose ΔT≤4 cutoff presupposes localization in Fock space. The authors explicitly state that a downward bend from L=12 onward cannot be excluded. Since the paper's central claim is persistence in the thermodynamic limit, this finite-size/prethermal ambiguity is load-bearing. I ask the authors to provide additional evidence against a prethermal plateau, for example long-time Krylov or tDMRG evolutions of ϵ(t) at J≤0.05 for L=12–16 showing no late-time growth toward ϵ=1, or a quantitative scaling analysis of δ that models a possible crossover to a downward bend.
  2. [Sec. II C, Eq. (8)] The Floquet-diagonal-ensemble formula for the infinite-time average assumes that no degeneracies occur in the quasienergy spectrum of the symmetric sector. The authors assert this is the case for nonvanishing U and J, but no proof is given; given the translation and parity symmetries, exact degeneracies are not obviously absent. If degeneracies exist, the infinite-time average of V does not reduce to Eq. (8) inside degenerate subspaces, and the quantities Vα, δ, and ϵ become dependent on the choice of eigenbasis. Please either prove the non-degeneracy for the parameter values used, or demonstrate numerically that the main conclusions are unchanged when the degenerate subspaces are diagonalized with a randomly chosen unitary.
  3. [Sec. II B, Sec. III A] The Hilbert-space truncation for L>10 fixes the maximum Fock-space distance ΔT≤4. This truncation is natural in a localized regime, but it cannot rule out a prethermal scenario in which the wave packet slowly spreads to larger Δ shells at times beyond the simulated window. The border projector Π_B bounds leakage within the truncated space, but only for the simulated time window. I request a convergence check in ΔT at fixed L (e.g., L=12–14) for the MBDL parameters, and, if feasible, an untruncated Krylov simulation at one intermediate size to verify that truncation does not suppress thermalization artificially.
minor comments (4)
  1. [Eq. (13)] The definition of δ uses ⟨log V⟩; if any Floquet state has Vα=0, the logarithm is singular. Since the uniform initial state is a single Fock state with V=0, the authors should state explicitly whether all Floquet states in the symmetric sector have strictly positive Vα, or restrict the average to Vα>0.
  2. [Abstract] The statement that the system does not generically heat up to infinite temperature is stronger than what is directly tested; the numerics explicitly consider only the uniform state and one excited product state. The all-spectrum distribution of Vα supports the claim, but it would be clearer to state that the conclusion is an inference from the Floquet-state distributions rather than from a survey of initial states.
  3. [Sec. IV] The interpretation that delocalization in Fock space implies ballistic propagation in real space is an indirect inference; the manuscript does not directly measure a real-space spreading observable such as density-density correlations. Please either add such a measurement or soften the wording.
  4. [Fig. 1 caption] The error bars for the truncated points are described as the average of the border projectior Π_B, but in the figure they are barely visible; please clarify their magnitude relative to the symbol size, or provide a table of values for the larger-L points.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the MBDL claim rests on direct numerical observables, not on fitted parameters or load-bearing self-citations.

full rationale

The paper's derivation chain is self-contained: the central MBDL conclusion is drawn from directly computed observables (epsilon in Eq. (6), delta in Eq. (13), overlap in Eq. (4), and entanglement entropy in Eq. (10)) evaluated by exact diagonalization, Krylov, and tDMRG, with no free parameter fitted to enforce the phase classification. The Hilbert-space truncation for L > 10 is checked by a border-projector expectation value rather than assumed, and the paper explicitly reports those error bars. Self-citations (Refs. [7,8,22,26]) supply standard diagnostics or prior related work, but the load-bearing evidence is the numerical scaling of the paper's own data. The paper also honestly flags its main limitation: "we emphasize that extrapolations to the thermodynamic limit should be taken with care," and "we cannot exclude from these data that the curves for J less than or similar to 0.05 start bending downwards from L = 12 onward but unfortunately our numerics cannot bring us further." These are honest finite-size caveats, not circular steps. No equation is defined in terms of the claimed result, no fitted quantity is renamed as a prediction, and no uniqueness claim is imported from the authors' prior work. Accordingly, the circularity score is 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The central claim is empirical, based on exact diagonalization and time evolution; no free parameter is fitted to force the conclusion. The main input assumptions are the validity of the Floquet diagonal ensemble, the Hilbert-space truncation, and the extrapolation to the thermodynamic limit. No new entities are introduced.

assumptions (5)
  • domain assumption The Floquet spectrum in the fully symmetric sector has no degeneracies, so the infinite-time average equals the Floquet diagonal ensemble (Eq. 8).
    Text states 'as it is the case when both U and J are nonvanishing and we have restricted to the fully-symmetric subspace', but no level statistics are shown to support this.
  • domain assumption For L>10, truncating the Hilbert space to Fock states with Delta(n0,n) <= Delta_T captures all physically relevant states.
    Used for all large-L data (Fig. 1, Fig. 4, Fig. 5). The truncation is chosen based on the initial state and could bias results toward localization; the border-projector expectation is used as a check, but it is not a rigorous bound.
  • domain assumption The small-J regime (J <= 0.05) is in the asymptotic large-L regime, so the upward bend of delta (Fig. 3) persists.
    This is the core extrapolation; the authors admit 'we cannot exclude from these data that the curves for J<=0.05 start bending downwards from L=12 onward'.
  • domain assumption Truncated Wigner approximation (TWA) captures the classical limit, meaning setting [n_hat, phi_hat] = 0 breaks quantum coherence sufficiently.
    In Sec. V and Appendix C, TWA is used to show classical diffusion; TWA is a semiclassical approximation, not an exact classical description.
  • domain assumption The uniform and excited initial states are representative of 'almost all initial states' because the Floquet-spectrum analysis shows all states violate ETH.
    The paper generalizes from two initial states and the V_alpha distribution, but does not test generic random initial states.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Many-body dynamical localization in the kicked Bose-Hubbard chain." pith.science (2026). https://pith.science/paper/VYMLYMBR

@misc{pith2026190803399,
  author       = {Pith},
  title        = {Pith review of: Many-body dynamical localization in the kicked Bose-Hubbard chain},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VYMLYMBR}},
  note         = {Machine review of arXiv:1908.03399}
}
read the original abstract

We provide evidence that a clean kicked Bose-Hubbard model exhibits a many-body dynamically localized phase. This phase shows ergodicity breaking up to the largest sizes we were able to consider. We argue that this property persists in the limit of large size. The Floquet states violate eigenstate thermalization and then the asymptotic value of local observables depends on the initial state and is not thermal. This implies that the system does not generically heat up to infinite temperature, for almost all the initial states. Differently from many-body localization here the entanglement entropy linearly increases in time. This increase corresponds to space-delocalized Floquet states which are nevertheless localized across specific subsectors of the Hilbert space: In this way the system is prevented from randomly exploring all the Hilbert space and does not thermalize.

Figures

Figures reproduced from arXiv: 1908.03399 by the authors.

Figure 2
Figure 2. FIG. 2. Upper panel, distribution of [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Dependence on [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. FIG. 4. Overlap dynamics in a MBDL case (upper panel – [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figures from the paper (16 more)
Figure 6
Figure 6. Figure 6: FIG. 6. Distribution of [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7. Floquet-state average [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 8
Figure 8. Figure 8: FIG. 8. Floquet-state standard deviation std( [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 10
Figure 10. Figure 10: FIG. 10. (Main figure) When a strong enough disorder is [PITH_FULL_IMAGE:figures/full_fig_p008_10.png]
Figure 11
Figure 11. Figure 11: FIG. 11. Distribution of [PITH_FULL_IMAGE:figures/full_fig_p009_11.png]
Figure 12
Figure 12. Figure 12: FIG. 12 [PITH_FULL_IMAGE:figures/full_fig_p010_12.png]
Figure 13
Figure 13. Figure 13: FIG. 13. Scaling of [PITH_FULL_IMAGE:figures/full_fig_p010_13.png]
Figure 14
Figure 14. Figure 14: FIG. 14. IPR [PITH_FULL_IMAGE:figures/full_fig_p011_14.png]
Figure 15
Figure 15. Figure 15: FIG. 15. Average log IPR against the average degeneracy [PITH_FULL_IMAGE:figures/full_fig_p011_15.png]
Figure 16
Figure 16. Figure 16: FIG. 16. Panel (a): for [PITH_FULL_IMAGE:figures/full_fig_p012_16.png]
Figure 17
Figure 17. Figure 17: FIG. 17. The diffusion coefficient [PITH_FULL_IMAGE:figures/full_fig_p012_17.png]
Figure 18
Figure 18. Figure 18: FIG. 18 [PITH_FULL_IMAGE:figures/full_fig_p014_18.png]
Figure 19
Figure 19. Figure 19: FIG. 19. Analog of Fig [PITH_FULL_IMAGE:figures/full_fig_p014_19.png]
Figure 20
Figure 20. Figure 20: FIG. 20 [PITH_FULL_IMAGE:figures/full_fig_p016_20.png]
Figure 21
Figure 21. Figure 21: FIG. 21. Size-scaling of the revival time for a MBDL case [PITH_FULL_IMAGE:figures/full_fig_p017_21.png]
Figure 22
Figure 22. Figure 22: FIG. 22. Overlap dynamics in a dynamically localized case [PITH_FULL_IMAGE:figures/full_fig_p017_22.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

71 extracted references · 44 canonical work pages

  1. [1]

    Fock basis

    is known to undergo a Kosterlitz-Thouless quantum phase transition in its ground state, from Mott insulator to superfluid [30, 31]. Recently the Kibble-Zurek scaling of excitations has been analyzed for a ramping through this quantum-phase-transition point [32]. The quench dynamics in a Bose-Hubbard model with static hopping has been numerically analyzed i...

  2. [2]

    J. M. Deutsch, Phys. Rev. A 43, 2046 (1991)

  3. [3]

    Srednicki, Phys

    M. Srednicki, Phys. Rev. E 50, 888 (1994)

  4. [4]

    Rigol, V

    M. Rigol, V. Dunjko, and M. Olshanii, Nature 452, 854 (2008)

  5. [5]

    H. Kim, T. N. Ikeda, and D. A. Huse, Phys. Rev. E 90, 052105 (2014)

  6. [6]

    D. A. Abanin, E. Altman, I. Bloch, and M. Serbyn, Rev. Mod. Phys. 91, 021001 (2019)

  7. [7]

    Eckardt, Rev

    A. Eckardt, Rev. Mod. Phys. 89, 011004 (2017)

  8. [8]

    Russomanno, A

    A. Russomanno, A. Silva, and G. E. Santoro, Phys. Rev. Lett. 109, 257201 (2012)

Show all 71 references
  1. [9]

    Russomanno, R

    A. Russomanno, R. Fazio, and G. E. Santoro, EPL 110, 37005 (2015)

  2. [10]

    Bukov, M

    M. Bukov, M. Heyl, D. A. Huse, and A. Polkovnikov, Phys. Rev. B 93, 155132 (2016)

  3. [11]

    Citro, R

    R. Citro, R. Citro, E. G. Dalla Torre, L. D’Alessio, A. Polkovnikov, M. Babadi, T. Oka, E. A. Demler , An- nals of Physics 360, 694 (2015)

  4. [12]

    T. Mori, T. Kuwahara, and K. Saito, Phys. Rev. Lett. 116, 120401 (2016)

  5. [13]

    Kuwahara, T

    T. Kuwahara, T. Mori, and K. Saito, Annals of Physics 367, 96 (2016)

  6. [14]

    Bukov, S

    M. Bukov, S. Gopalakrishnan, M. Knap, and E. Demler, Phys. Rev. Lett. 115, 205301 (2015)

  7. [15]

    S. A. Weidinger and M. Knap, Scientific Reports7, 45382 (2017)

  8. [16]

    M. V. Berry, Regular and Irregular Motion, in Topics in Nonlinear Mechanics, edited by S. Jorna 46, pp. 16–120, Am. Inst. Ph., 1978

  9. [17]

    Lichtenberg and M

    A. Lichtenberg and M. Lieberman, Regular and Chaotic Dynamics Applied Mathematical Sciences, 2 nd ed. (Springer New York, 1989)

  10. [18]

    Chirikov and V

    B. Chirikov and V. Vecheslavov, Journal of Statistical Physics 71, 243 (1993)

  11. [19]

    Lazarides, A

    A. Lazarides, A. Das, and R. Moessner, Phys. Rev. E 90, 012110 (2014)

  12. [20]

    D’Alessio and M

    L. D’Alessio and M. Rigol, Phys. Rev. X 4, 041048 (2014)

  13. [21]

    B. V. Chirikov, Time-dependent quantum systems, in Chaos and quantum mechanics, edited by M.-J.Giannoni, A. Voros, and J. Zinn-Justin, Les Houches Lecture Se- ries Vol. 52, p. 443–545, Elsevier Sci. Publ., Amsterdam, 1991

  14. [22]

    F. I. G. Casati, B.V. Chirikov and J. Ford, in Stochastic Behaviour in classical and Quantum Hamiltonian Sys- tems, edited by G. Casati and J. Ford, , Lecture Notes in Physics Vol. 93, Springer N. Y., 1979

  15. [23]

    Notarnicola, F

    S. Notarnicola, F. Iemini, D. Rossini, R. Fazio, A. Silva, and A. Russomanno, Phys. Rev. E 97, 022202 (2018)

  16. [24]

    E. B. Rozenbaum and V. Galitski, Phys. Rev. B 95, 064303 (2017)

  17. [25]

    Rylands, E

    C. Rylands, E. Rozenbaum, V. Galitski, and R. Konik, arXiv , 1904.09473 (2019). 17 4 5 6 7 8 9 10 3000 3500 4000 4500 5000 5500 6000 6500 7000 FIG. 21. Size-scaling of the revival time for a MBDL case (J = 0.01) and excited initial state with initial state|ψ02(0)⟩. Notice the ...

  18. [26]

    Parameswaran, A

    S. Parameswaran, A. Potter, and R. Vasseur, Annalen der Physik 529, 1600302 (2017)

  19. [27]

    Russomanno, G

    A. Russomanno, G. E. Santoro, and R. Fazio, Journal of Statistical Mechanics: Theory and Experiment 2016, 073101 (2016)

  20. [28]

    Alba and P

    V. Alba and P. Calabrese, PNAS 114, 7947-7951 (2017)

  21. [29]

    Alba and P

    V. Alba and P. Calabrese, SciPost Phys. 4, 17 (2018)

  22. [30]

    Calabrese and J

    P. Calabrese and J. Cardy, Journal of Statistical Me- chanics: Theory and Experiment 2005, P04010 (2005)

  23. [31]

    M. P. A. Fisher, P. B. Weichman, G. Grinstein, and D. S. Fisher, Phys. Rev. B 40, 546 (1989)

  24. [32]

    Elstner and H

    N. Elstner and H. Monien, arXiv , 9905367 (1999)

  25. [33]

    Gardas, J

    B. Gardas, J. Dziarmaga, and W. H. Zurek, Phys. Rev. B 95, 104306 (2017)

  26. [34]

    Kollath, A

    C. Kollath, A. M. L¨ auchli, and E. Altman, Phys. Rev. Lett. 98, 180601 (2007)

  27. [35]

    Kollath, G

    C. Kollath, G. Roux, G. Biroli, and A. M. L¨ auchli, Jour- nal of Statistical Mechanics: Theory and Experiment 2010, P08011 (2010)

  28. [36]

    Rispoli, A

    M. Rispoli, A. Lukin, R. Schittko, S. Kim, M. E, Tai, J. L´ eonard, M. Greiner, Nature573, 385 (2019)

  29. [37]

    D. N. Page Phys. Rev. Lett., 71 1291-1294 (1993)

  30. [38]

    Lukin, M

    A. Lukin, M. Rispoli, R. Schittko, M. E. Tai, A. M. Kauf- man, S. Choi, V. Khemani, J. L´ eonard, M. Greiner, Sci- ence 364, 256 (2019)

  31. [39]

    For L = 11 we average over Nreal Floquet states randomly chosen among the 16158 ones

    For L≤ 10 we average over all the Floquet states. For L = 11 we average over Nreal Floquet states randomly chosen among the 16158 ones. Nreal = 620 for J = 0.15 andNreal∼ 4000 for the other cases. We have checked for smaller system sizes that the results given by this pro- ced...

  32. [40]

    The physical reason behind the absence of parity effect in dynamical localization is that in this case the quantum correlations extend over a length scale smaller than the system size, opposite to the thermalizing case where the correlations extend over all the system being the...

  33. [41]

    R. B. Sidje, ACM Trans. Math. Softw. 24, 130 (1998)

  34. [42]

    Schollwock, Ann

    U. Schollwock, Ann. Phys. 326, 96 (2011)

  35. [43]

    ITensor Library (version 2.0.11) http://itensor.org

  36. [44]

    Ponte, A

    P. Ponte, A. Chandran, Z. Papi´ c, and D. A. Abanin, Annals of Physics 353, 196 (2015)

  37. [45]

    For VT =∞ we either use its exact finite-size value, com- puted directly as the trace of ˆV in the symmetric sub- space, or forL> 10 we approximateVT =∞/L≃ν2 which becomes exact in the L→∞ limit, as can be seen by ex- changing the fixed-number ensemble with a grand canon- ical o...

  38. [46]

    Sambe, Phys

    H. Sambe, Phys. Rev. A 7, 2203 (1973)

  39. [47]

    J. H. Shirley, Phys. Rev. 138, B979 (1965)

  40. [48]

    J. T. Edwards and D. J. Thouless, J. Phys. C 5, 807 (1972)

  41. [49]

    M. A. Nielsen and I. L. Chuang, Quantum Computation and Quantum Information: 10th Anniversary Edition , 10th ed. (Cambridge University Press, New York, NY, USA, 2011)

  42. [50]

    Peres, Phys

    A. Peres, Phys. Rev. A 30, 504 (1984)

  43. [51]

    De Chiara, S

    G. De Chiara, S. Montangero, P. Calabrese and R. Fazio, Journal of Statistical Mechanics: Theory and Experi- ment 2006, P03001 (2006)

  44. [52]

    ˇZnidariˇ c, T

    M. ˇZnidariˇ c, T. Prosen, and P. Prelovˇ sek, Phys. Rev. B 77, 064426 (2008)

  45. [53]

    J. H. Bardarson, F. Pollmann, and J. E. Moore, Phys. Rev. Lett. 109, 017202 (2012)

  46. [54]

    Serbyn, Z

    M. Serbyn, Z. Papi´ c, and D. A. Abanin, Phys. Rev. Lett. 110, 260601 (2013)

  47. [55]

    Singh, J

    R. Singh, J. H. Bardarson, and F. Pollmann, New Jour- nal of Physics 18, 023046 (2016)

  48. [56]

    J. A. Kj¨ all, J. H. Bardarson, and F. Pollmann, Phys. Rev. Lett. 113, 107204 (2014)

  49. [57]

    D. J. Luitz, N. Laflorencie, and F. Alet, Phys. Rev. B 91, 081103(R) (2015)

  50. [58]

    D. A. Abanin, W. D. Roeck, and F. Huveneers, Annals of Physics 372, 1 (2016)

  51. [59]

    Ponte, Z

    P. Ponte, Z. Papi´ c, F. Huveneers, and D. A. Abanin, Phys. Rev. Lett. 114, 140401 (2015)

  52. [60]

    F. M. Surace, A. Russomanno, M. Dalmonte, A. Silva, R. Fazio, and F. Iemini, Phys. Rev. B 99, 104303 (2019)

  53. [61]

    Sierant, D

    P. Sierant, D. Delande, and J. Zakrzewski, Phys. Rev. A 95, 021601(R) (2017)

  54. [62]

    Sierant, D

    P. Sierant, D. Delande, and J. Zakrzewski, Acta Physica Polonica A 132, 1707 (2017). 18

  55. [63]

    Sierant and J

    P. Sierant and J. Zakrzewski, New J. Phys. 20, 043032 (2018)

  56. [64]

    G. B. Mbeng, Localizzazione a molti corpi in una catena di bosoni fortemente interagenti , Mas- ter Thesis, Universit` a di Pisa (2015), available at https://etd.adm.unipi.it/t/etd-09252015-163432/

  57. [65]

    We discuss in Appendix B some features of a case with J intermediate between these two conditions

  58. [66]

    Polkovnikov, Annals of Physics 325, 1790 (2010)

    A. Polkovnikov, Annals of Physics 325, 1790 (2010)

  59. [67]

    Chandran, I

    A. Chandran, I. H. Kim, G. Vidal, and D. A. Abanin, Phys. Rev. B 91, 085425 (2015)

  60. [68]

    However one should be careful in this analogy since, unlike a classical probability distribution function, the Wigner distribution is not always positive, i. e. it may have regions inψ space where W (ψ)< 0. This is why it is often referred to as quasi-distribution

  61. [69]

    Gerry and P

    C. Gerry and P. Knight, Introductory Quantum Optics (Cambridge University Press, 2004)

  62. [70]

    Frenkel and B

    D. Frenkel and B. Smit, Understanding molecular simu- lation: from algorithms to applications Vol. 1 (Elsevier, 2001)

  63. [71]

    Efron, The jackknife, the bootstrap, and other resam- pling plans Vol

    B. Efron, The jackknife, the bootstrap, and other resam- pling plans Vol. 38 (Siam, 1982)

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.