Pith. sign in

REVIEW 1 major objections 6 minor 31 references

Random drift zeroes essential spectrum in Brox diffusions

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · glm-5.2

2026-07-09 11:39 UTC pith:DLOTFCNZ

load-bearing objection Solid paper proving new spectral results for Brox diffusions in random environments; one minor fixable gap in Lemma 5.1(ii), otherwise clean. the 1 major comments →

arxiv 2607.07410 v1 pith:DLOTFCNZ submitted 2026-07-08 math.PR math.SP

Essential spectrum for Brox-type diffusion processes

Yuichi Shiozawa , Jian Wang This is my paper
classification math.PR math.SP
keywords randomalmostalphadiffusionessentialfieldspotentialsprocesses
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This paper studies the spectral properties of Brox diffusion processes—particles diffusing through random potential landscapes—in two types of random media: stationary Gaussian fields and semi-selfsimilar Lévy fields. The central claim is that in both settings, the essential spectral bottom of the diffusion operator equals zero almost surely, meaning the associated Markov semigroup is noncompact. The authors prove this for multi-dimensional Gaussian environments with sublinearly growing covariance (including fractional Brownian fields) and for multi-dimensional semi-selfsimilar Lévy environments with scaling index α in (1,2), as well as for one-dimensional symmetric α-stable processes with any α in (0,2). The key mechanism is a chain of implications: sample path behavior of the random potential determines the volume growth rate of the random reference measure μ_W(dx) = e^{-W(x)}dx, and that volume growth rate, via known spectral criteria for regular Dirichlet forms, pins down the essential spectral bottom. Crucially, the authors also show that random Lévy drift perturbations can destroy the compactness of semigroups generated by deterministic power-law potentials ±|x|^δ when 0 < δ ≤ 1/α, even when the unperturbed semigroup is compact—demonstrating that random fluctuations can reverse deterministic confinement properties.

Core claim

The paper establishes a unified framework linking the sample path behavior of random environmental potentials to the spectral characteristics of the resulting diffusion semigroup. The core technical discovery is that almost sure volume growth estimates for the random measure μ_W, derived from Gaussian extremal theory and Lévy scaling ergodicity, directly determine the essential spectral bottom. For Gaussian environments with sublinearly growing covariance, the volume growth rate (1/R)log μ_W(B(0,R)) tends to zero almost surely, forcing λ_ess = 0. For semi-selfsimilar Lévy environments, the strong mixing property of the scaling transform (Lemma 4.2) combined with the Birkhoff ergodic theorem驱

What carries the argument

The central machinery is a three-step pipeline: (a) derive almost sure sample path bounds for the random potential W(x) using Gaussian extremal estimates (Kôno-type results) or Lévy process maximal inequalities; (b) translate these bounds into volume growth estimates for μ_W(B(0,R)) via direct integration; (c) apply spectral criteria for regular Dirichlet forms (specifically, volume growth upper bounds on the essential spectral bottom from Notarantonio's theorem and Pinsky's one-dimensional noncompactness criterion) to conclude λ_ess = 0. For the Lévy case, the strong mixing of the scaling transform T_r (Lemma 4.2) is the key ergodic-theoretic input enabling the Birkhoff theorem application.

Load-bearing premise

The proof that the essential spectral bottom equals zero for semi-selfsimilar Lévy environments relies on the strong mixing property of the scaling transform combined with the assumption that negative Lévy increments occur with positive probability. If the mixing property does not hold with sufficient strength for a given semi-selfsimilar process, the Birkhoff ergodic theorem cannot be applied to establish the volume growth divergence that underpins the spectral conclusion.

What would settle it

Construct a semi-selfsimilar Lévy environment where the scaling transform fails to be strongly mixing, or where negative increments have probability zero, and show that the volume growth of μ_W(B(0,R)) remains bounded—this would break the chain from sample paths to λ_ess = 0.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Share X Bluesky LinkedIn Reddit HN

If this is right

  • For fractional Brownian fields with any Hurst parameter H in (0,1) and any dimension d, the Brox diffusion semigroup is almost surely noncompact with zero essential spectral bottom, settling the spectral question for this entire class.
  • When α < 1, the threshold 1/α exceeds 1, so there exist deterministic potentials ±|x|^δ with δ in (1, 1/α] that generate compact semigroups but become noncompact under random Lévy perturbation—random fluctuations can override deterministic confinement.
  • The volume-growth-to-spectrum pipeline is stated by the authors as generalizable to other random potential models beyond Gaussian and Lévy types.
  • The regime δ > 1/α remains open: the paper cannot determine whether compactness survives random perturbation for large potential exponents.

Where Pith is reading between the lines

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

  • The zero essential spectral bottom for Gaussian environments with sublinearly growing covariance suggests that any random potential whose fluctuations grow slower than linearly at infinity is insufficient to confine the diffusion in a spectral sense, regardless of the fine structure of correlations.
  • The mixing property of the scaling transform is the load-bearing ergodic input for the Lévy case; if one could construct a semi-selfsimilar Lévy process whose scaling transform fails to be strongly mixing, the volume growth divergence argument would break, and the spectral conclusion could fail.
  • The perturbation result (Proposition 5.2) hints at a phase transition at δ = 1/α between regimes where random drift dominates and where deterministic confinement dominates, but the paper leaves the critical regime unresolved.
  • The framework could potentially extend to other self-similar random structures (e.g., multifractional Brownian motion or anisotropic Lévy fields) if analogous sample path bounds and ergodic properties can be established.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

1 major / 6 minor

Summary. This paper studies the essential spectrum and compactness of Markov semigroups associated with multi-dimensional Brox diffusion processes in random environments. Two types of random media are considered: stationary Gaussian random fields and semi-selfsimilar Lévy random fields. For Gaussian environments satisfying general stationary covariance growth conditions, the authors prove that the associated semigroup is almost surely noncompact, and if the covariance function grows sublinearly at infinity, the bottom of the essential spectrum vanishes almost surely. For multi-dimensional semi-selfsimilar Lévy environments with scaling index α∈(1,2), they show that the essential spectral bottom equals zero almost surely for arbitrary dimension d≥1. The same conclusion is established for one-dimensional symmetric α-stable Lévy processes with any α∈(0,2). Finally, in one dimension, the authors demonstrate that random Lévy drift perturbations can destroy the compactness of semigroups induced by deterministic power-law potentials ±|x|^δ when 0<δ≤1/α, even when the unperturbed semigroup is compact. The proofs proceed by establishing almost sure volume growth properties of the random reference measure and then applying known spectral criteria (Notarantonio, Pinsky).

Significance. The paper makes a solid contribution by extending spectral analysis of Brox-type diffusions to multi-dimensional random environments, a setting where prior work focused primarily on recurrence/transience. The unified framework linking sample path behaviors of random potentials to spectral characteristics via volume growth estimates is a clear methodological strength. The result that random Lévy drift can destroy compactness of confining deterministic potentials (Proposition 5.2) is a notable qualitative finding. The proofs are detailed and self-contained where it matters: Lemma 4.2 provides a complete proof of the strong mixing property of the scaling transform using characteristic functions, and the Borel-Cantelli arguments and Birkhoff ergodic theorem applications are correctly executed. The restriction to α∈(1,2) in Theorem 4.1 is clearly stated and the symmetric stable case in d=1 covers α∈(0,2) separately via Proposition 4.6.

major comments (1)
  1. Lemma 5.1(ii), proof of (5.1): The proof states that for 0<δ≤1/α and c>r_0, the inequality cr_0^{n/α} ≥ r_0^{(n+1)δ} holds for large n. For δ<1/α this holds for any c>0. However, for δ=1/α, the condition becomes c ≥ r_0^{1/α}. When α<1 (so 1/α>1), r_0^{1/α}>r_0, so c>r_0 is insufficient. The fix is straightforward: choose c>r_0^{1/α} (possible since Q(m(1)>c)>0 for any c by Lemma 5.1(i)). This does not affect Proposition 5.2 or the main qualitative result, but the stated condition in the proof should be corrected.
minor comments (6)
  1. Proposition 5.2 does not explicitly restate the conditions Q(V(1)>0)>0 and Q(V(1)<0)>0 from Lemma 5.1, though these are needed. Adding a brief statement of these assumptions in the proposition would improve clarity.
  2. In the proof of Proposition 4.5(2), the asymptotic equivalence for the integral of e^{|t|^{1/α}(log|t|)^p} is stated without justification. A brief comment on how this follows (e.g., Laplace-type asymptotics) would help the reader.
  3. The reference [6] (Chen and Wang) is cited as a preprint arXiv:2509.08559 but does not appear to be used in the paper. Either cite it or remove it.
  4. In the proof of Lemma 4.4, the transition from discrete points to all x≥r_0^N via right continuity is mentioned but could be stated more explicitly for the negative half-line, since V is defined with different continuity properties on (-∞,0].
  5. The abstract states results for 'semi-selfsimilar Lévy environments' but Proposition 4.6 covers the r_0=1 (stable) case separately. A brief remark in the abstract or introduction noting that the stable case in d=1 is covered by a different argument would improve accuracy.
  6. In equation (4.9), the inequality μ_W(B(0,R)) ≤ (∫_{-R}^{R} e^{-V(t)} dt)^d uses the inclusion B(0,R) ⊂ [-R,R]^d. This is correct but the factorization into a d-th power follows because W(x) = Σ V(x_k) with independent V(x_k) across coordinates. A brief note clarifying this would help the reader.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for a careful reading and for identifying a genuine gap in the proof of Lemma 5.1(ii). The referee's observation is correct: when δ = 1/α and α < 1, the condition c > r_0 is insufficient to guarantee cr_0^{n/α} ≥ r_0^{(n+1)δ} for large n. We will revise the proof accordingly.

read point-by-point responses

  1. Referee: Lemma 5.1(ii), proof of (5.1): The proof states that for 0<δ≤1/α and c>r_0, the inequality cr_0^{n/α} ≥ r_0^{(n+1)δ} holds for large n. For δ<1/α this holds for any c>0. However, for δ=1/α, the condition becomes c ≥ r_0^{1/α}. When α<1 (so 1/α>1), r_0^{1/α}>r_0, so c>r_0 is insufficient. The fix is straightforward: choose c>r_0^{1/α} (possible since Q(m(1)>c)>0 for any c by Lemma 5.1(i)). This does not affect Proposition 5.2 or the main qualitative result, but the stated condition in the proof should be corrected.

    Authors: The referee is entirely correct, and we are grateful for this careful observation. When δ = 1/α, the inequality cr_0^{n/α} ≥ r_0^{(n+1)δ} = r_0^{n/α + 1/α} reduces to c ≥ r_0^{1/α}. Since r_0 > 1 and 1/α > 1 when α < 1, we have r_0^{1/α} > r_0, so the condition c > r_0 stated in the proof is indeed insufficient in this case. The fix is exactly as the referee describes: we replace the condition c > r_0 with c > r_0^{1/α}. This is permissible because Lemma 5.1(i) guarantees Q(m(1) > c) > 0 for any c > 0, so the Birkhoff ergodic theorem argument in (5.5) and the subsequent divergence conclusion remain valid. The case δ < 1/α requires no change since any c > 0 suffices. We emphasize, as the referee notes, that this correction does not affect Proposition 5.2 or the main qualitative result. We will update the proof in the revised manuscript to state the condition c > r_0^{1/α} (or more precisely, c > r_0^{δ} for general δ ∈ (0, 1/α], which specializes to c > r_0^{1/α} at the boundary). revision: yes

Circularity Check

0 steps flagged

No significant circularity. The self-citation to [28] (Shiozawa) provides a general spectral criterion also attributed to [23] (Notarantonio, external), and the paper's central results—volume growth estimates—are derived independently from first principles.

full rationale

The paper's derivation chain is self-contained against external benchmarks. The central results (Theorems 3.2, 4.1; Propositions 4.6, 5.2) are obtained by independently proving almost-sure volume growth estimates for the random reference measure μ_W (Propositions 3.3, 4.5; Lemma 5.1) and then applying a general spectral criterion for regular Dirichlet forms. The spectral criterion is cited as '[23, Theorem 1] (see also [28, Theorem 3.2])'—[23] is by Notarantonio (external author, arXiv:math/9806002), while [28] is by Shiozawa (one of the current authors). Since the same criterion is available from an external source, the self-citation is not load-bearing in a circular sense. The volume growth estimates themselves are proved from scratch using Fernique's theorem, Borel-Cantelli lemmas, ergodic theory (Lemma 4.2 on mixing is proved in full via characteristic functions), and sample path bounds (Lemma 4.4 is proved using Chebyshev and Borel-Cantelli). Lemma 4.3 and Lemma 5.1(i) cite [19, Proposition 2.1] by Kusuoka–Takahashi–Tamura (external). Proposition 4.6 uses [4, Theorem 1] (Bertoin–Yor) and [26, Theorem 2] (Pinsky), both external. Proposition 5.2 uses [26, Theorem 2] (Pinsky, external). No step in the derivation chain reduces to its own inputs by construction. The one minor self-citation ([28]) is a general tool applied to new settings, not a restatement of the target result.

Axiom & Free-Parameter Ledger

3 free parameters · 4 axioms · 0 invented entities

The paper introduces no new physical entities or particles. The free parameters (α, δ, H) are standard parameters of the stochastic processes and potentials considered. The axioms are either domain assumptions (Assumption 3.1 on covariance structure) or standard mathematical results from spectral theory and probability.

free parameters (3)
  • α (scaling index) = α∈(1,2) for Theorem 4.1; α∈(0,2) for Proposition 4.6
    The scaling index of the semi-selfsimilar Lévy process is a parameter of the model, not fitted to data but chosen as a condition on the class of processes considered.
  • δ (potential exponent) = 0<δ≤1/α
    The exponent of the deterministic power-law potential, constrained in the perturbation result.
  • H (Hurst parameter) = H∈(0,1)
    Parameter for fractional Brownian fields in Example 3.4.
axioms (4)
  • domain assumption Assumption 3.1: Stationary covariance structure with specific growth and regularity conditions on γ and σ_i.
    Section 3, Assumption 3.1. This is a modeling assumption on the Gaussian environment, not a standard mathematical axiom.
  • standard math The scaling transform T_r is strongly mixing (Lemma 4.2).
    Section 4.2, Lemma 4.2. Proved in the paper using characteristic functions and independent increments.
  • standard math Volume growth criterion for essential spectrum (Notarantonio [23, Theorem 1] / Shiozawa [28, Theorem 3.2]).
    Used in proofs of Theorems 3.2 and 4.1. A known result from spectral theory of Dirichlet forms.
  • standard math Pinsky's noncompactness criterion for 1D diffusion operators [26, Theorem 2].
    Used in Proposition 4.6 and Proposition 5.2. A standard result in one-dimensional spectral theory.

pith-pipeline@v1.1.0-glm · 21660 in / 2614 out tokens · 496872 ms · 2026-07-09T11:39:00.135077+00:00 · methodology

0 comments
Cite this review

Pith. "Pith review of Essential spectrum for Brox-type diffusion processes." pith.science (2026). https://pith.science/paper/DLOTFCNZ

@misc{pith2026260707410,
  author       = {Pith},
  title        = {Pith review of: Essential spectrum for Brox-type diffusion processes},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DLOTFCNZ}},
  note         = {Machine review of arXiv:2607.07410}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

This paper investigates the essential spectrum and compactness property of Markov semigroups generated by multi-dimensional Brox diffusion processes under two types of random media: stationary Gaussian random fields and semi-selfsimilar L\'evy random fields. For Gaussian environments satisfying general stationary covariance growth conditions, we prove that the associated semigroup is almost surely noncompact; if the covariance function grows sublinearly at infinity, then the bottom of the essential spectrum vanishes almost surely, with fractional Brownian fields as a concrete example. For multi-dimensional semi-selfsimilar L\'evy environments with scaling index $\alpha\in(1,2)$, we show that the essential spectral bottom equals zero almost surely for arbitrary space dimension $d\ge1$, and the same conclusion holds for one-dimensional symmetric $\alpha$-stable L\'evy processes with any $\alpha\in(0,2)$. Furthermore, we study one-dimensional diffusion operators perturbed by random L\'evy drift. When $0<\delta\le 1/\alpha$, random environmental fluctuations can destroy the compactness of semigroups induced by deterministic power-law potentials $\pm|x|^\delta$, even if the unperturbed semigroup is compact. The analysis relies on almost sure volume growth estimates for the random reference measure induced by environmental potentials, sample path asymptotics of random fields, ergodic theory of scaling transforms, and spectral criteria for regular Dirichlet forms. A unified framework linking sample path behaviors of random potentials to spectral characteristics of diffusion semigroups is established, and several open problems for large potential exponents are stated.

discussion (0)

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

Reference graph

Works this paper leans on

31 extracted references · 31 canonical work pages · 2 internal anchors

  1. [1]

    P. W. Anderson, Absence of diffusion in certain random lattices,Physical Review,109 (1958), 1492–1505

    work page 1958

  2. [2]

    Avellaneda and A.J

    M. Avellaneda and A.J. Majda, Mathematical models with exact renormalization for tur- bulent transport,Comm. Math. Phys.131(1990), 381–429

    work page 1990

  3. [3]

    Bakry, I

    D. Bakry, I. Gentil and M. Ledoux,Analysis and Geometry of Markov Diffusion Operators, Springer, Cham, 2014

    work page 2014

  4. [4]

    Bertoin and M

    J. Bertoin and M. Yor, Exponential functionals of L´ evy processes,Probab. Surv.2(2005), 191–212

    work page 2005

  5. [5]

    Brox, A one-dimensional diffusion process in a Wiener medium,Ann

    Th. Brox, A one-dimensional diffusion process in a Wiener medium,Ann. Probab.14 (1986), 1206–1218

    work page 1986

  6. [6]

    Quenched and annealed heat kernel estimates for Brox's diffusion

    X. Chen and J. Wang, Quenched and annealed heat kernel estimates for Brox’s diffusion, preprint, arXiv:2509.08559

    work page internal anchor Pith review Pith/arXiv arXiv

  7. [7]

    G. S. Choi, Criteria for recurrence and transience of semistable processes,Nagoya Math. J.134(1994), 91–106

    work page 1994

  8. [8]

    Fazekas and Z

    I. Fazekas and Z. Rychlik, Almost sure limit theorems for semi-selfsimilar processes,Probab. Math. Statist.25(2005), 241–255

    work page 2005

  9. [9]

    Fukushima, S

    M. Fukushima, S. Nakao and M. Takeda, On Dirichlet forms with random data–recurrence and homogenization, Stochastic processes - Mathematics and Physics, II (Bielefeld, 1985), Lecture Notes in Math.1250, 87–97, Springer-Verlag, Berlin, 1987

    work page 1985

  10. [10]

    Fukushima, Y

    M. Fukushima, Y. Oshima and M. Takeda,Dirichlet Forms and Symmetric Markov Pro- cesses, Second revised and extended edition, Walter de Gruyter, Berlin, 2011

    work page 2011

  11. [11]

    Havlin and D

    S. Havlin and D. Ben-Avraham, Diffusion in disordered media,Advances in Physics51 (2002), 187–292

    work page 2002

  12. [12]

    Itˆ o, On the ergodicity of a certain stationary process.Proc

    K. Itˆ o, On the ergodicity of a certain stationary process.Proc. Imp. Acad. Tokyo20(1944), 54–55. 19

    work page 1944

  13. [13]

    Khintchine, Zwei S¨ atze ¨ uber stochastische Prozesse mit stabilen Verteilungen,Rec

    A. Khintchine, Zwei S¨ atze ¨ uber stochastische Prozesse mit stabilen Verteilungen,Rec. Math. Moscou,3(1938), 577–584

    work page 1938

  14. [14]

    Kim and S

    D. Kim and S. Kusuoka, Recurrence of direct products of diffusion processes in random media having zero potentials,Electron. J. Probab.25(2020), 1–18

    work page 2020

  15. [15]

    Kˆ ono, Asymptotic behavior of sample functions of Gaussian random fields,J

    N. Kˆ ono, Asymptotic behavior of sample functions of Gaussian random fields,J. Math. Kyoto Univ.15(1975), 671–707

    work page 1975

  16. [16]

    K¨ uhn and R

    F. K¨ uhn and R. L. Schilling, Maximal inequalities and some applications,Probab. Surv. 20(2023), 382–485

    work page 2023

  17. [17]

    Kusuoka, H

    S. Kusuoka, H. Takahashi and Y. Tamura, Recurrence of the Brownian motion in multidi- mensional semi-selfsimilar environments and Gaussian environments,Potential Anal.43 (2015), 695–705

    work page 2015

  18. [18]

    Kusuoka, H

    S. Kusuoka, H. Takahashi and Y. Tamura, Topics on multi-dimensional Brox’s diffusions, RIMS Kˆ okyˆ uroku Bessatsu,B59, Research Institute for Mathematical Sciences (RIMS), Kyoto, 2016, pp. 31–43

    work page 2016

  19. [19]

    Kusuoka, H

    S. Kusuoka, H. Takahashi and Y. Tamura, Recurrence and transience properties of multi- dimensional diffusion processes in selfsimilar and semi-selfsimilar random environments, Electron. Commun. Probab.22(2017), 1–11

    work page 2017

  20. [20]

    Lifshits,Lectures on Gaussian Processes, SpringerBriefs Math., Springer, Heidelberg, 2012

    M. Lifshits,Lectures on Gaussian Processes, SpringerBriefs Math., Springer, Heidelberg, 2012

    work page 2012

  21. [21]

    Maejima and K

    M. Maejima and K. Sato, Semi-selfsimilar processes,J. Theoret. Probab.12(1999), 347– 373

    work page 1999

  22. [22]

    Mathieu, Zero white noise limit through Dirichlet forms, with application to diffusions in a random medium,Probab

    P. Mathieu, Zero white noise limit through Dirichlet forms, with application to diffusions in a random medium,Probab. Theory Related Fields99(1994), 549–580

    work page 1994

  23. [23]

    Growth and spectrum of diffusions

    L. Notarantonio, Growth and spectrum of diffusions, preprint, arXiv:math/9806002

    work page internal anchor Pith review Pith/arXiv arXiv

  24. [24]

    Orey, Growth rate of certain Gaussian processes, Proceedings of the Sixth Berkeley Symposium on Mathematical Statistics and Probability (Univ

    S. Orey, Growth rate of certain Gaussian processes, Proceedings of the Sixth Berkeley Symposium on Mathematical Statistics and Probability (Univ. California, Berkeley, Calif., 1970/1971), Vol. II: Probability theory, pp. 443–451, University of California Press, Berke- ley, CA, 1972

    work page 1970

  25. [25]

    Pavliotis,Stochastic Processes and Applications, Springer, New York, 2014

    G. Pavliotis,Stochastic Processes and Applications, Springer, New York, 2014

    work page 2014

  26. [26]

    R. G. Pinsky, Explicit and almost explicit spectral calculations for diffusion operators,J. Funct. Anal.256(2009), 3279–3312

    work page 2009

  27. [27]

    Sato,L´ evy Processes and Infinitely Divisible Distributions, Revised edition, Cambridge University Press, Cambridge, 2013

    K. Sato,L´ evy Processes and Infinitely Divisible Distributions, Revised edition, Cambridge University Press, Cambridge, 2013

    work page 2013

  28. [28]

    Shiozawa, Volume growth, big jump, and essential spectrum for regular Dirichlet forms, Integral Equations Operator Theory97(2025), article number 31, 35pp

    Y. Shiozawa, Volume growth, big jump, and essential spectrum for regular Dirichlet forms, Integral Equations Operator Theory97(2025), article number 31, 35pp

    work page 2025

  29. [29]

    Tanaka, Recurrence of a diffusion process in a multidimensional Brownian environment, Proc

    H. Tanaka, Recurrence of a diffusion process in a multidimensional Brownian environment, Proc. Japan Acad. Ser. A. Math. Sci.69(1993), 377–381

    work page 1993

  30. [30]

    Walters,An Introduction to Ergodic Theory, Springer-Verlag, New York-Berlin, 1982

    P. Walters,An Introduction to Ergodic Theory, Springer-Verlag, New York-Berlin, 1982

    work page 1982

  31. [31]

    Wang,Functional Inequalities, Markov Processes and Spectral Theory, Science Press, Beijing, 2005

    F.-Y. Wang,Functional Inequalities, Markov Processes and Spectral Theory, Science Press, Beijing, 2005. 20

    work page 2005