Pith. sign in

REVIEW 4 major objections 5 minor 34 references

Discontinuous observables as an obstruction for small essential spectral radius

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read This paper shows that for piecewise smooth expanding maps, any Banach space of observables that contains indicator functions of intervals with controlled norm has essential spectral radius at least $1/\Theta_\infty(1-s)$, and that the…

desk verdict The L∞ barrier is genuinely lowered to L^1 spaces, but Theorem D's proof needs a serious fix before the lower bound is established. read the letter →

arxiv 2506.07613 v1 pith:VWRKPHTP submitted 2025-06-09 math.DS

Oliver Butterley , Daniel Smania This is my paper
classification math.DS MSC 37A0537A2537C30
keywords transferoperatoressentialspectralradiuspiecewiseexpandingmapsBesovspacesdiscontinuousobservablesinfinite-dimensionaleigenspacesgapatomicdecomposition
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

This paper shows that the essential spectral radius of a transfer operator on a Banach space of observables cannot be made arbitrarily small if the space contains simple discontinuities, namely indicator functions of intervals with controlled norm. For piecewise $C^{1+\beta}$ expanding maps preserving Lebesgue measure, any such space automatically contains the Besov space $B^s_{1,1}$, and the essential spectral radius is at least $1/\Theta_\infty(1-s)$, where $\Theta_\infty(1-s)$ encodes the exponential growth rate of inverse-derivative sums. Moreover, every complex number with modulus below this threshold is an eigenvalue of the transfer operator with an infinite-dimensional eigenspace. This matters because many natural function spaces, such as Besov and Sobolev spaces, are not embedded in $L^\infty$, so earlier results did not apply; this work extends the obstruction to unbounded observables and to natural norms.

What carries the argument

The engine of the proof is the eigenfunction series $h_z = \sum_{\ell\ge 0} z^\ell \psi\circ T^\ell$, built from a non-zero observable $\psi = \varphi_1 \mathbf{1}_{J_1} - \varphi_2 \mathbf{1}_{J_2}$ satisfying $L\psi=0$. For each $z$ with $|z|$ below the threshold, the series converges in the Besov space $B^s_{1,1}(I)$, defined by atomic decomposition into scaled interval indicators, using the estimate $\|\psi\circ T^k\|_{B^s_{1,1}} \le C\Theta_k(1-s)$, and then $B^s_{1,1}$ embeds continuously into $B$ by the hypothesis on indicator norms. Since the terms $\psi\circ T^\ell$ are mutually orthogonal in $L^2$, $h_z$ is non-zero and satisfies $Lh_z = z h_z$; varying $\psi$ produces infinitely many independent eigenfunctions. The quantity $\Theta_\infty(1-s) = \lim_k (\sum_i (\theta^k_i)^{1-s})^{1/k}$ measures the exponential growth of the inverse-derivative sums and sets the radius of the forced disc.

What would settle it

Exhibit one piecewise $C^{1+\beta}$ expanding map $T$ with Lebesgue-invariant measure and one Banach space $B$ satisfying $\|\mathbf{1}_J\|_B \le C|J|^{1-s}$ with $s<\beta$, for which the only $\psi$ of the form $\varphi_1\mathbf{1}_{J_1} - \varphi_2\mathbf{1}_{J_2}$ with $L\psi=0$ is zero, or for which $r_{\mathrm{ess}}(L,B) < 1/\Theta_\infty(1-s)$; a concrete place to look is a non-Markovian piecewise expanding map where the two intervals must be chosen ad hoc and the two-term cancellation equation $L\psi=0$ can be solved explicitly.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central discovery is that the mere possibility of a discontinuity, an indicator function $\mathbf{1}_J$ of an interval, already forces the essential spectrum of the transfer operator to be large, independent of the detailed choice of Banach space. Theorem A states that if $B$ embeds continuously in $L^1(I)$, contains every $\mathbf{1}_J$ with $\|\mathbf{1}_J\|_B \le C|J|^{1-s}$ for some $s\in(0,1)$, and $T$ is piecewise $C^{1+\beta}$ with $\beta>s$, then the essential spectral radius satisfies $r_{\mathrm{ess}}(L,B) \ge 1/\Theta_\infty(1-s)$, and the open disc $|\lambda|<1/\Theta_\infty(1-s)$ consists entirely of eigenvalues with infinite-dimensional eigenspaces. The same mechanism yields the lower bound $1/k$ for piecewise linear maps with $k$ branches under the weaker bound $\|\mathbf{1}_J\|_B \le C$, and for Markov maps under a topological-pressure condition. Theorem D shows that for any Banach space with a natural norm, either the uniform-bound case giving $1/k$ or the scaling-bound case giving $1/\Theta_\infty(1-s)$ must occur, so a spectral gap cannot be achieved by moving to such spaces.

Load-bearing premise

The proof assumes without proof that, for the given piecewise expanding map $T$, there exists a non-zero coboundary observable $\psi$ made of two pieces, $\varphi_1$ on one interval minus $\varphi_2$ on another, with $L\psi=0$; the entire eigenfunction construction and the lower bound depend on this existence.

Editorial extensions

If this is right

  • For any Banach space satisfying the indicator-norm condition with $s<\beta$, the Besov space $B^s_{1,1}$ is continuously embedded, so lower bounds for $B^s$ carry over to $B$.
  • The essential spectral radius of $L$ on such a space is at least $1/\Theta_\infty(1-s)$, and every $|\lambda|<1/\Theta_\infty(1-s)$ is an eigenvalue with infinite-dimensional eigenspace, so the essential spectrum contains a disc of that radius.
  • For piecewise linear expanding maps with $k$ branches, the weaker uniform indicator bound $\|\mathbf{1}_J\|_B \le C$ already forces $r_{\mathrm{ess}}(L,B) \ge 1/k$.
  • For Markovian $C^{r+1}$ maps satisfying $P_{\mathrm{top}}(-(r+1)\log|DT|) < 1/k$, the same $1/k$ lower bound holds under uniform indicator bounds.
  • For any Banach space with a natural norm containing indicators and on which $L$ has a spectral gap, Theorem D forces the indicator norms to scale like $|P|^{-t}$ with $t\in[0,1)$, and then $r_{\mathrm{ess}}(L,B) \ge 1/k$ in case $t=0$ or $\ge 1/\Theta_\infty(1-s)$ in case $s=1+t\in(0,1)$.

Reading between the lines

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

  • If Theorem A is correct, any attempt to build a Banach space with a spectral gap for a piecewise expanding map must either exclude even the simplest step-function observables or accept a minimal essential spectral radius tied to the derivative growth $\Theta_\infty(1-s)$, making the search for optimal spaces a quantitative problem.
  • The infinite-dimensional eigenspaces in the disc $|\lambda|<1/\Theta_\infty(1-s)$ imply that decay-of-correlations rates faster than that threshold are impossible on spaces containing discontinuous observables, a direct spectral interpretation that the paper only sketches in Remark 2.1.8.
  • A natural extension beyond the interval is that the same construction should yield analogous lower bounds in higher-dimensional piecewise expanding settings whenever coboundary observables supported on small cells with controlled variation exist, since the obstruction appears intrinsic to discontinuities rather than to dimension one.
  • For numerical or rigorous spectral computations on Besov or Sobolev spaces of piecewise expanding maps, these bounds predict an observable disc of essential spectrum of radius at least $1/\Theta_\infty(1-s)$; a computed smaller essential radius would signal that the chosen norm is not natural or that the indicator bound fails.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper studies lower bounds on the essential spectral radius of transfer operators associated with piecewise expanding interval maps, acting on Banach spaces that contain characteristic functions of intervals. The main theorems (A–D) assert that, under rather mild norm estimates on interval indicators, the essential spectral radius has an intrinsic lower bound, and that certain small complex numbers are eigenvalues with infinite-dimensional eigenspaces. The proofs combine a construction of eigenfunctions from zero-observables with Besov-space embedding estimates, and apply these to "natural" Banach spaces in Theorem D.

Significance. If correct, the results give a broad and useful obstruction principle: for many Banach function spaces that admit discontinuous observables, the essential spectral radius cannot be arbitrarily small. This complements existing work for L^p and BV-type spaces and covers unbounded observables, which is a genuine improvement over earlier L^∞-based results. The paper also proposes a natural notion of "natural" norm with a homogeneity degree, which appears to be a valuable organizing concept for studying transfer operators on Besov and Sobolev spaces. The lower bounds are consistent with known examples and no contradiction with existing results is apparent.

major comments (4)
  1. [Theorem D, final paragraph (p. 14)] Case II is concluded by setting s = 1 + t_max(I) and saying 'the remaining conclusions follow from Theorem A', but the proof never verifies the hypothesis s < β (with β = r in the notation of Theorem A). The intermediate Claims II and III only give 1 + t_max(I) ≥ β_pos > 0, which is a lower bound on s, not the required upper bound s < r. When s ≥ r, Proposition 3.2.16, which is used inside the proof of Theorem A with p = 1/β, cannot be applied, and the eigenfunctions h_z constructed in Theorem A are not shown to belong to B^s_{1,1}. Therefore the proof of the Case II lower bound is incomplete as written.
  2. [Proof of Theorem A, Section 4 (p. 8)] The proof begins by asserting the existence of a non-zero ψ ∈ L^∞(I) with Lψ = 0, and later asserts that one can take ψ = φ_1 1_{J_1} − φ_2 1_{J_2} with φ_i of bounded 1/β-variation. No proof or explicit reference is given for this existence step. Since every subsequent eigenvalue construction and the lower bound on the essential spectral radius depends on the availability of such zero-observables, this is a load-bearing gap. The authors should either prove the existence for the stated class of piecewise C^{1+β} maps or state it as a separate lemma with a complete proof or a precise citation.
  3. [Proof of Theorem C, Claims B and D (pp. 10–12)] The claim in Claim B that the image of h_{z,n} is a Cantor set (up to a countable set), and the subsequent use in Claim D to conclude h_{z,n} + w_{z,n} ≠ 0, are only sketched. The argument that the image is a Cantor set relies on the conformal expanding map G_n and the full-branch assumption, but the justification that h_{z,n} agrees with the function ĥ_{z,n} whose image is the Cantor set is not fully spelled out. Since Claim D is essential for producing eigenvalues of L^n and hence for the lower bound r_ess(L,B) ≥ 1/k, this sketch should be expanded into a rigorous argument.
  4. [Theorem D, Claim IV (p. 14)] The intervals Q_i are defined as Q_i = [x_i, x_{i+1}] with x_0 = 0 and |Q_i| = θ^i. Then Q_0 = [0,1], Q_1 = [1, 1+θ], and so on, so the union [0, x_{k+1}] is not contained in I = [0,1]. The estimates from Claim I only apply to intervals inside I, so the proof of t_max(I) ≤ 0 does not work as written. This appears to be a typo (e.g., |Q_i| = (1−θ)θ^i would be needed), but as stated it breaks the argument.
minor comments (5)
  1. [Section 1, paragraph after the definition of T (p. 3)] There is a missing term in the sentence 'Then the transfer operator L with respect to the is a bounded operator acting on L^1(I)'; presumably 'with respect to the Lebesgue measure' is intended.
  2. [Section 1.1 (p. 3)] The notation alternates between T and f (e.g., '|D f^k(x)|' and '|D T(x)|'). Using a single symbol for the map would improve readability.
  3. [Remark 1.1.1 (p. 3)] The inequality Θ_∞(β) ≤ #P_1 lim_k (sup_x 1/|D f^k(x)|^β)^{1/k} appears to require the sup to be finite and the limit to exist; the authors should clarify the precise hypotheses under which this is stated.
  4. [Section 3.2 (p. 7)] The norm on B^s_{1,1} is denoted |φ|_{B^s_{1,1}(I)}, which conflicts with the usual use of |·| for Lebesgue measure earlier in the paper; using ||·|| would avoid ambiguity.
  5. [Theorem C, Claim A (p. 10)] The decomposition of B into a finite-dimensional subspace and a closed invariant subspace with spectral radius < |z| is invoked, but the existence of such a decomposition for the non-quasi-compact operator on the Banach space B requires some argument or reference; the paper currently does not justify this step.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: lower bounds are derived from the indicator-norm hypotheses and external results; the flagged issues in Theorem D are proof gaps, not circular reductions.

full rationale

The derivation chain is self-contained in the relevant sense. Theorem A's lower bound is obtained by constructing eigenfunctions h_z = Σ z^ℓ ψ∘T^ℓ from a zero observable ψ and estimating their Besov norm via Proposition 3.2.16 and the indicator bound (2.1.2); the conclusion r_ess ≥ 1/Θ∞(1−s) is not assumed in these hypotheses. The Besov embedding B^s_{1,1}⊂B follows directly from the atomic representation (3.2.14) and (2.1.2). The cited orthogonality of F_ℓ (de Lima–Smania [15]) is an external published result used only to guarantee h_z≠0, and it does not contain the target lower bound. Theorems B and C rely on direct series estimates and on Collet–Isola's independent spectral result. Theorem D obtains the interval-scaling exponent from naturality and then applies Theorem A or C. I also flag three non-circular gaps, weighed separately: (i) the existence of ψ=φ_1 1_{J_1}−φ_2 1_{J_2} with Lψ=0 is asserted without proof in the proof of Theorem A; (ii) Case II of Theorem D invokes Theorem A without verifying s<β, since only s∈(0,1) and r>0 are established; (iii) Claim IV uses intervals [0,x_{k+1}] that may leave I=[0,1], so the norm bounds for intervals inside I may not apply. These passages are places where support is missing or the proof is incomplete, but none of them makes a prediction equal to its input by construction.

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

No new entities or fitted parameters appear. The paper's findings depend on standard transfer-operator facts and on the asserted existence of zero-mode observables for piecewise expanding maps.

assumptions (4)
  • domain assumption Existence of zero-mode observables: for each piecewise expanding map T with invariant Lebesgue measure, there is an infinite-dimensional space of non-zero functions ψ of the form φ_1 1_{J_1} − φ_2 1_{J_2} with bounded 1/β-variation pieces such that Lψ = 0.
    Asserted in the proofs of Theorem A and B, not proved. The construction of all eigenfunctions in the paper depends on it.
  • standard math Collet-Isola formula: for a C^{r+1} Markov expanding map, r_ess(L, C^r(I)) = exp(P_top(−(r+1) log|DT|)).
    Used in Theorem C to fix the radius within which eigenvalues must exist; taken from [13].
  • standard math The atomic Besov space B^s_{1,1}(I) from (3.2.14) coincides with the classical Besov space and satisfies the dyadic approximation estimate of Proposition 3.2.16.
    Established in de Souza and Smania; a proof sketch for the proposition is provided in Section 3.2.
  • standard math Spectral gap on B implies exponential decay of L^n on the zero-mean subspace, i.e., ||L^n a||_B ≤ C λ^n ||a||_B for ∫ a dm=0.
    Used in Theorem D, Claim II, as a standard consequence of quasi-compactness with a simple eigenvalue at 1.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Discontinuous observables as an obstruction for small essential spectral radius." pith.science (2026). https://pith.science/paper/VWRKPHTP

@misc{pith2026250607613,
  author       = {Pith},
  title        = {Pith review of: Discontinuous observables as an obstruction for small essential spectral radius},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VWRKPHTP}},
  note         = {Machine review of arXiv:2506.07613}
}
read the original abstract

We show that for a very wide class of Banach spaces of functions on [0,1] there are intrinsic lower bounds for the essential spectral radius of the transfer operator associated to piecewise smooth expanding maps. The class of Banach spaces studied includes any reasonable space which permits discontinuities.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

34 extracted references · 34 canonical work pages

  1. [1]

    Transfer operators and atomic decomposition, 2020

    Alexander Arbieto and Daniel Smania. Transfer operators and atomic decomposition, 2020

    work page 2020

  2. [2]

    Baladi and G

    V . Baladi and G. Keller. Zeta functions and transfer operators for piecewise monotone transformations. Comm. Math. Phys., 127(3):459–477, 1990

    work page 1990

  3. [3]

    World Scientific Publishing Co., Inc., River Edge, NJ, 2000

    Viviane Baladi.Positive transfer operators and decay of correlations, volume 16 ofAdvanced Series in Non- linear Dynamics. World Scientific Publishing Co., Inc., River Edge, NJ, 2000

    work page 2000

  4. [4]

    Viviane Baladi.Dynamical zeta functions and dynamical determinants for hyperbolic maps, volume 68 ofErgebnisse der Mathematik und ihrer Grenzgebiete. 3. Folge. A Series of Modern Surveys in Mathematics [Results in Mathematics and Related Areas. 3rd Series. A Series of Modern Surveys in Mathematics]. Springer, Cham, 2018. A functional approach

    work page 2018

  5. [5]

    Anisotropic Hölder and Sobolev spaces for hyperbolic diffeomorphisms

    Viviane Baladi and Masato Tsujii. Anisotropic Hölder and Sobolev spaces for hyperbolic diffeomorphisms. Ann. Inst. Fourier (Grenoble), 57(1):127–154, 2007

    work page 2007

  6. [6]

    Dynamical determinants and spectrum for hyperbolic diffeomorphisms

    Viviane Baladi and Masato Tsujii. Dynamical determinants and spectrum for hyperbolic diffeomorphisms. InGeometric and probabilistic structures in dynamics, volume 469 ofContemp. Math., pages 29–68. Amer. Math. Soc., Providence, RI, 2008

    work page 2008

  7. [7]

    Ruelle-Perron-Frobenius spectrum for Anosov maps.Nonlinearity, 15(6):1905–1973, 2002

    Michael Blank, Gerhard Keller, and Carlangelo Liverani. Ruelle-Perron-Frobenius spectrum for Anosov maps.Nonlinearity, 15(6):1905–1973, 2002

    work page 1905

  8. [8]

    Springer-Verlag, Berlin, revised edition, 2008

    Rufus Bowen.Equilibrium states and the ergodic theory of Anosov diffeomorphisms, volume 470 ofLecture Notes in Mathematics. Springer-Verlag, Berlin, revised edition, 2008. Edited by Chazottes, Jean-René. With a preface by David Ruelle

    work page 2008

Show all 34 references
  1. [9]

    Discontinuities cause essential spectrum on surfaces.Annales Henri Poincaré, 2024

    Oliver Butterley, Giovanni Canestrari, and Roberto Castorrini. Discontinuities cause essential spectrum on surfaces.Annales Henri Poincaré, 2024

    2024

  2. [10]

    Discontinuities cause essential spectrum.Comm

    Oliver Butterley, Giovanni Canestrari, and Sakshi Jain. Discontinuities cause essential spectrum.Comm. Math. Phys., 398(2):627–653, 2023

    2023

  3. [11]

    Locating Ruelle-Pollicott resonances.Nonlin- earity, 35(1):513–566, 2022

    Oliver Butterley, Niloofar Kiamari, and Carlangelo Liverani. Locating Ruelle-Pollicott resonances.Nonlin- earity, 35(1):513–566, 2022

    2022

  4. [12]

    American Mathematical Society, Providence, RI, 2006

    Nikolai Chernov and Roberto Markarian.Chaotic billiards, volume 127 ofMathematical Surveys and Monographs. American Mathematical Society, Providence, RI, 2006

    2006

  5. [13]

    On the essential spectrum of the transfer operator for expanding Markov maps.Comm

    Pierre Collet and Stefano Isola. On the essential spectrum of the transfer operator for expanding Markov maps.Comm. Math. Phys., 139(3):551–557, 1991

    1991

  6. [14]

    I. P . Cornfeld, S. V . Fomin, and Ya. G. Sinai.Ergodic theory, volume 245 ofGrundlehren der mathematis- chen Wissenschaften [Fundamental Principles of Mathematical Sciences]. Springer-Verlag, New York, 1982. Translated from the Russian by A. B. Sosinski˘i. 16 OLIVER BUTTERLEY...

    1982

  7. [15]

    On infinitely cohomologous to zero observables.Ergodic Theory Dynam

    Amanda de Lima and Daniel Smania. On infinitely cohomologous to zero observables.Ergodic Theory Dynam. Systems, 33(2):375–399, 2013

    2013

  8. [16]

    The atomic decomposition of Besov-Bergman-Lipschitz spaces.Proceedings of the American Mathematical Society, 94(4):682–686, 1985

    Geraldo Soares de Souza. The atomic decomposition of Besov-Bergman-Lipschitz spaces.Proceedings of the American Mathematical Society, 94(4):682–686, 1985

    1985

  9. [17]

    Geraldo Soares De Souza, Richard O’Neil, and G. Sampson. Several characterizations for the special atom spaces with applications.Rev. Mat. Iberoamericana, 2(3):333–355, 1986

    1986

  10. [18]

    Ruelle spectrum of linear pseudo-Anosov maps

    Frédéric Faure, Sébastien Gouëzel, and Erwan Lanneau. Ruelle spectrum of linear pseudo-Anosov maps. J. Éc. polytech. Math., 6:811–877, 2019

    2019

  11. [19]

    Lagarias, and Bjorn Poonen

    Leopold Flatto, Jeffrey C. Lagarias, and Bjorn Poonen. The zeta function of the beta transformation.Er- godic Theory Dynam. Systems, 14(2):237–266, 1994

    1994

  12. [20]

    Banach spaces adapted to Anosov systems.Ergodic Theory Dynam

    Sébastien Gouëzel and Carlangelo Liverani. Banach spaces adapted to Anosov systems.Ergodic Theory Dynam. Systems, 26(1):189–217, 2006

    2006

  13. [21]

    V . M. Gundlach and Y. Latushkin. A sharp formula for the essential spectral radius of the Ruelle transfer operator on smooth and Hölder spaces.Ergodic Theory Dynam. Systems, 23(1):175–191, 2003

    2003

  14. [22]

    G. Keller. Markov extensions, zeta functions, and Fredholm theory for piecewise invertible dynamical sys- tems.Trans. Amer . Math. Soc., 314(2):433–497, 1989

    1989

  15. [23]

    On the rate of convergence to equilibrium in one-dimensional systems.Comm

    Gerhard Keller. On the rate of convergence to equilibrium in one-dimensional systems.Comm. Math. Phys., 96(2):181–193, 1984

    1984

  16. [24]

    Stability of the spectrum for transfer operators.Ann

    Gerhard Keller and Carlangelo Liverani. Stability of the spectrum for transfer operators.Ann. Scuola Norm. Sup. Pisa Cl. Sci. (4), 28(1):141–152, 1999

    1999

  17. [25]

    Eigenfunctions for smooth expanding circle maps.Nonlinearity, 17(5):1723–1730, 2004

    Gerhard Keller and Hans Henrik Rugh. Eigenfunctions for smooth expanding circle maps.Nonlinearity, 17(5):1723–1730, 2004

    2004

  18. [26]

    Lasota and James A

    A. Lasota and James A. Yorke. On the existence of invariant measures for piecewise monotonic transfor- mations.Trans. Amer . Math. Soc., 186:481–488 (1974), 1973

    1974

  19. [27]

    Rigorous numerical investigation of the statistical properties of piecewise expanding maps

    Carlangelo Liverani. Rigorous numerical investigation of the statistical properties of piecewise expanding maps. A feasibility study.Nonlinearity, 14(3):463–490, 2001

    2001

  20. [28]

    Invariant measures and their properties

    Carlangelo Liverani. Invariant measures and their properties. A functional analytic point of view. InDy- namical systems. Part II, Pubbl. Cent. Ric. Mat. Ennio Giorgi, pages 185–237. Scuola Norm. Sup., Pisa, 2003

    2003

  21. [29]

    Spectra of expanding maps on Besov spaces.Discrete Contin

    Yushi Nakano and Shota Sakamoto. Spectra of expanding maps on Besov spaces.Discrete Contin. Dyn. Syst., 39(4):1779–1797, 2019

    2019

  22. [30]

    Zeta functions and the periodic orbit structure of hyperbolic dynamics

    William Parry and Mark Pollicott. Zeta functions and the periodic orbit structure of hyperbolic dynamics. Astérisque, 187-188:284, 1990

    1990

  23. [31]

    The thermodynamic formalism for expanding maps.Comm

    David Ruelle. The thermodynamic formalism for expanding maps.Comm. Math. Phys., 125(2):239–262, 1989

    1989

  24. [32]

    Transfer operators, atomic decomposition and the Bestiary

    Daniel Smania. Transfer operators, atomic decomposition and the Bestiary. Arxiv preprint 1903.06976, 2021

    1903 arXiv

  25. [33]

    Besov-ish spaces through atomic decomposition.Anal

    Daniel Smania. Besov-ish spaces through atomic decomposition.Anal. PDE, 15(1):123–174, 2022

    2022

  26. [34]

    A spectral gap for transfer operators of piecewise expanding maps.Discrete Contin

    Damien Thomine. A spectral gap for transfer operators of piecewise expanding maps.Discrete Contin. Dyn. Syst., 30(3):917–944, 2011. DISCONTINUOUS OBSERVABLES AS AN OBSTRUCTION FOR SMALL ESSENTIAL SPECTRAL RADIUS 17 (OLIVERBUTTERLEY) DEPARTMENT OFMATHEMATICS, UNIVERSITY OFROMET...

    2011

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.