pith. sign in

arxiv: 2509.16016 · v3 · submitted 2025-09-19 · 🧮 math.SP · cs.CC· cs.NA· math.DS· math.NA

Residual SCI Upper Bounds And Lower Witnesses For Koopman Approximate Point Spectra In L^p For 1<p<infty: Extended Version

Pith reviewed 2026-05-18 15:51 UTC · model grok-4.3

classification 🧮 math.SP cs.CCcs.NAmath.DSmath.NA
keywords Koopman operatorapproximate point spectrumpseudospectrumspectral computationL^p spacesresidual boundsmeasure-preserving maps
0
0 comments X

The pith

Continuous finite-dimensional dictionaries and tagged quadrature residuals yield upper bounds on approximate point spectra of Koopman operators in L^p spaces.

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

The paper establishes that approximate point spectral sets for bounded Koopman operators on L^p spaces can be approximated from above in a residual, computable manner. It separates the regularized approximate point ε-pseudospectrum from its closed counterpart and shows how to obtain these upper bounds, along with bounds on the approximate point spectrum itself, when the underlying map belongs to one of four classes. The input to the procedure is pointwise access to the map on a compact metric space equipped with a finite Borel measure, and the output is measured in the Hausdorff metric on compact subsets of the complex plane. A reader would care because the approach turns an abstract spectral question into a concrete numerical task that works uniformly across continuous nonsingular maps, maps with controlled modulus of continuity, measure-preserving maps, and maps satisfying both conditions.

Core claim

Using continuous finite-dimensional dictionaries and tagged quadrature residuals, we prove SCI upper bounds for R_ap,ε(T), C_ap,ε(T), and σ_ap on four natural classes of maps: continuous nonsingular maps, maps with a prescribed modulus of continuity, measure-preserving maps, and maps satisfying both measure preservation and a prescribed modulus.

What carries the argument

Continuous finite-dimensional dictionaries paired with tagged quadrature residuals, which produce explicit upper bounds on the Hausdorff distance to the target spectral sets.

If this is right

  • Upper bounds hold for both the regularized and closed approximate point ε-pseudospectra on continuous nonsingular maps.
  • The same residual method supplies upper bounds when the map satisfies a prescribed modulus of continuity.
  • Measure-preserving maps admit computable upper bounds on the same spectral sets.
  • Maps that are both measure-preserving and modulus-controlled receive the tightest residual upper bounds among the four classes.
  • Lower witnesses accompany the upper bounds to certify the quality of the approximation.

Where Pith is reading between the lines

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

  • The residual technique may adapt to other classes of operators that admit finite-dimensional dictionary approximations on L^p spaces.
  • Numerical experiments on standard dynamical systems such as circle rotations could test how quickly the Hausdorff error decreases with dictionary dimension.
  • The distinction between regularized and closed pseudospectra suggests different stability properties under perturbation of the map.

Load-bearing premise

The map is available only through point evaluations and the finite Borel measure makes the Koopman operator bounded on L^p for 1 < p < ∞.

What would settle it

For a concrete measure-preserving map whose approximate point spectrum is known exactly, compute the dictionary-based residual upper bound and check whether the true spectrum lies inside the claimed Hausdorff neighborhood.

read the original abstract

We study residual computation of approximate point spectral sets of bounded Koopman operators $\mathcal K_F$ on $L^p(\mathcal X,\omega)$, $1<p<\infty$, where $\mathcal X$ is a compact metric space and $\omega$ is a finite Borel measure. The input is the underlying map $F : \mathcal X \to \mathcal X$, accessed through point evaluations, and the output metric is the Hausdorff metric on non-empty compact subsets of $\mathbb C$. For a bounded operator $T$, we distinguish the regularized approximate point $\varepsilon$-pseudospectrum $R_{\mathrm{ap},\varepsilon}(T)$ from the closed approximate point $\varepsilon$-pseudospectrum $C_{\mathrm{ap},\varepsilon}(T)$. The latter is the direct closed lower-norm analogue of the approximate point $\varepsilon$-pseudospectrum used in the $L^2$ Koopman SCI theory. Using continuous finite-dimensional dictionaries and tagged quadrature residuals, we prove SCI upper bounds for $R_{\mathrm{ap},\varepsilon}(T)$, $C_{\mathrm{ap},\varepsilon}(T)$, and $\sigma_{\mathrm{ap}}$ on four natural classes of maps: continuous nonsingular maps, maps with a prescribed modulus of continuity, measure-preserving maps, and maps satisfying both measure preservation and a prescribed modulus.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The manuscript develops residual SCI upper bounds for the regularized approximate point ε-pseudospectrum R_ap,ε(T), the closed approximate point ε-pseudospectrum C_ap,ε(T), and the approximate point spectrum σ_ap of bounded Koopman operators K_F acting on L^p(X,ω) with 1<p<∞. The map F is accessed only via point evaluations on the compact metric space X, and bounds are derived in the Hausdorff metric using continuous finite-dimensional dictionaries together with tagged quadrature residuals. The results are stated for four classes: continuous nonsingular maps, maps with a prescribed modulus of continuity, measure-preserving maps, and maps that are both measure-preserving and have a prescribed modulus of continuity.

Significance. If the derivations are correct, the work extends the existing L^2 Koopman SCI theory to the L^p setting for 1<p<∞ and supplies an explicit residual-based computational pathway for spectral-set approximation. The separation of regularized and closed pseudospectra is a useful technical distinction. The approach via dictionaries and quadrature residuals is constructive and potentially reproducible, which strengthens the computational relevance of the claims.

major comments (1)
  1. [Abstract and setting and notation] Abstract and 'setting and notation' section: the central claims for the class of continuous nonsingular maps presuppose that K_F is a bounded operator on L^p(X,ω). Nonsingularity of F with respect to ω is necessary for K_F to map L^p into itself but is not sufficient for boundedness; boundedness additionally requires that the Radon-Nikodym derivative d(ω ∘ F^{-1})/dω lies in L^∞(ω). The manuscript states only that ω 'allows the Koopman operator to be well-defined and bounded' without imposing or verifying this L^∞ condition. If the condition fails for some continuous nonsingular F, then K_F is unbounded and the asserted SCI upper bounds for R_ap,ε(T) and C_ap,ε(T) on this class do not apply. This assumption is load-bearing for one of the four main classes.
minor comments (1)
  1. The abstract introduces 'tagged quadrature residuals' without a short inline definition or pointer to the precise definition in the text; adding one sentence would improve immediate readability.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful reading and constructive comments on the manuscript. We address the major comment below and agree that an explicit clarification will improve the presentation.

read point-by-point responses
  1. Referee: [Abstract and setting and notation] Abstract and 'setting and notation' section: the central claims for the class of continuous nonsingular maps presuppose that K_F is a bounded operator on L^p(X,ω). Nonsingularity of F with respect to ω is necessary for K_F to map L^p into itself but is not sufficient for boundedness; boundedness additionally requires that the Radon-Nikodym derivative d(ω ∘ F^{-1})/dω lies in L^∞(ω). The manuscript states only that ω 'allows the Koopman operator to be well-defined and bounded' without imposing or verifying this L^∞ condition. If the condition fails for some continuous nonsingular F, then K_F is unbounded and the asserted SCI upper bounds for R_ap,ε(T) and C_ap,ε(T) on this class do not apply. This assumption is load-bearing for one of the four main classes.

    Authors: We thank the referee for highlighting this important technical detail. We agree that nonsingularity of F ensures K_F maps L^p into itself but does not by itself guarantee boundedness; the Radon-Nikodym derivative d(ω ∘ F^{-1})/dω must additionally lie in L^∞(ω). The manuscript assumes from the outset that K_F is a bounded operator on L^p(X,ω), with the phrase 'ω allows the Koopman operator to be well-defined and bounded' intended to encompass all conditions necessary for boundedness. To remove any ambiguity and strengthen the exposition, we will revise the setting and notation section to state explicitly that, for the class of continuous nonsingular maps, the results require both nonsingularity and the essential boundedness of the Radon-Nikodym derivative. This clarification will be added without altering the statements or proofs of the main results. revision: yes

Circularity Check

0 steps flagged

Derivation is self-contained with no circular reductions

full rationale

The paper establishes SCI upper bounds for R_ap,ε(T), C_ap,ε(T), and σ_ap by constructing continuous finite-dimensional dictionaries and tagged quadrature residuals on the four specified classes of maps. These constructions operate directly from point evaluations of F and the assumption that ω makes K_F bounded on L^p, without any step that defines the target spectral sets in terms of themselves or renames fitted quantities as predictions. No load-bearing self-citations, uniqueness theorems imported from prior author work, or ansatzes smuggled via citation are required for the central claims. The derivation chain therefore remains independent of its outputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard background results from functional analysis and measure theory together with the specific approximation constructions; no free parameters, ad-hoc axioms, or new postulated entities are visible in the abstract.

axioms (1)
  • standard math Koopman operator K_F induced by a map F is bounded on L^p(X,ω) for 1<p<∞ when ω is a finite Borel measure on compact metric space X.
    Invoked implicitly when the setting is introduced and the operator is called bounded.

pith-pipeline@v0.9.0 · 5785 in / 1455 out tokens · 75369 ms · 2026-05-18T15:51:49.602981+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. From Witness-Space Sharpness To Family-Pointwise Exactness For The Solvability Complexity Index

    math.LO 2026-04 unverdicted novelty 6.0

    Witness-space sharpness coincides with worst-case exactness for SCI but is strictly weaker than family-pointwise exactness in general, with upgrade theorems, a decoder-regular transport preorder, and examples from int...

  2. From Witness-Space Sharpness To Family-Pointwise Exactness For The Solvability Complexity Index

    math.LO 2026-04 unverdicted novelty 6.0

    Formalizes a trichotomy of exactness notions for SCI on families, proves witness-space sharpness equals worst-case exactness but is strictly weaker than family-pointwise exactness, supplies upgrade theorems, and illus...

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages · cited by 1 Pith paper

  1. [1]

    Good measures on cantor space

    Ethan Akin. Good measures on cantor space. Transactions of the American Mathematical Society , 357(7):2681--2722, 2005

  2. [2]

    Ergodic theory, dynamic mode decomposition, and computation of spectral properties of the koopman operator

    Hassan Arbabi and Igor Mezic. Ergodic theory, dynamic mode decomposition, and computation of spectral properties of the koopman operator. SIAM Journal on Applied Dynamical Systems , 16(4):2096--2126, 2017

  3. [3]

    Contractive projections in lp spaces

    Tsuyoshi And \^o . Contractive projections in lp spaces. Pacific Journal of Mathematics , 17(3):391--405, 1966

  4. [4]

    Computing spectra--on the solvability complexity index hierarchy and towers of algorithms

    Jonathan Ben-Artzi, Matthew J Colbrook, Anders C Hansen, Olavi Nevanlinna, and Markus Seidel. Computing spectra--on the solvability complexity index hierarchy and towers of algorithms. arXiv preprint arXiv:1508.03280 , 2015

  5. [5]

    New barriers in complexity theory: On the solvability complexity index and the towers of algorithms

    Jonathan Ben-Artzi, Anders C Hansen, Olavi Nevanlinna, and Markus Seidel. New barriers in complexity theory: On the solvability complexity index and the towers of algorithms. Comptes Rendus Mathematique , 353(10):931--936, 2015

  6. [6]

    Positive transfer operators and decay of correlations , volume 16 of Advanced Series in Nonlinear Dynamics

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

  7. [7]

    arXiv preprint arXiv:2102.12086 , year=

    Steven L Brunton, Marko Budi s i \'c , Eurika Kaiser, and J Nathan Kutz. Modern koopman theory for dynamical systems. arXiv preprint arXiv:2102.12086 , 2021

  8. [8]

    Orthogonality in linear metric spaces

    Garrett Birkhoff. Orthogonality in linear metric spaces. Duke Math. J. , 1(2):169--172, 1935

  9. [9]

    Applied koopmanism

    Marko Budi s i \'c , Ryan Mohr, and Igor Mezi \'c . Applied koopmanism. Chaos: An Interdisciplinary Journal of Nonlinear Science , 22(4), 2012

  10. [10]

    The foundations of spectral computations via the solvability complexity index hierarchy

    Matthew J Colbrook and Anders C Hansen. The foundations of spectral computations via the solvability complexity index hierarchy. Journal of the European Mathematical Society , 25(12):4639--4718, 2022

  11. [11]

    Limits and powers of koopman learning

    Matthew J Colbrook, Igor Mezi \'c , and Alexei Stepanenko. Limits and powers of koopman learning. arXiv preprint arXiv:2407.06312 , 2024

  12. [12]

    Operator theoretic aspects of ergodic theory , volume 272

    Tanja Eisner, B \'a lint Farkas, Markus Haase, and Rainer Nagel. Operator theoretic aspects of ergodic theory , volume 272. Springer, 2015

  13. [13]

    A counterexample to the approximation problem in B anach spaces

    Per Enflo. A counterexample to the approximation problem in B anach spaces. Acta Math. , 130:309--317, 1973

  14. [14]

    A new class of unbalanced haar wavelets that form an unconditional basis for lp on general measure spaces

    Maria Girardi and Wim Sweldens. A new class of unbalanced haar wavelets that form an unconditional basis for lp on general measure spaces. Journal of Fourier Analysis and Applications , 3(4):457--474, 1997

  15. [15]

    On the solvability complexity index, the -pseudospectrum and approximations of spectra of operators

    Anders Hansen. On the solvability complexity index, the -pseudospectrum and approximations of spectra of operators. Journal of the American Mathematical Society , 24(1):81--124, 2011

  16. [16]

    Robert C. James. Orthogonality and linear functionals in normed linear spaces. Trans. Amer. Math. Soc. , 61:265--292, 1947

  17. [17]

    Johnson and Joram Lindenstrauss

    William B. Johnson and Joram Lindenstrauss. Basic concepts in the geometry of B anach spaces. In Handbook of the geometry of B anach spaces, V ol. I , pages 1--84. North-Holland, Amsterdam, 2001

  18. [18]

    Generalized bounded variation and applications to piecewise monotonic transformations

    Gerhard Keller. Generalized bounded variation and applications to piecewise monotonic transformations. Z. Wahrsch. Verw. Gebiete , 69(3):461--478, 1985

  19. [19]

    Hamiltonian systems and transformation in hilbert space

    Bernard O Koopman. Hamiltonian systems and transformation in hilbert space. Proceedings of the National Academy of Sciences , 17(5):315--318, 1931

  20. [20]

    Data-driven spectral analysis of the koopman operator

    Milan Korda, Mihai Putinar, and Igor Mezi \'c . Data-driven spectral analysis of the koopman operator. Applied and Computational Harmonic Analysis , 48(2):599--629, 2020

  21. [21]

    On lipschitz partitions of unity and the assouad--nagata dimension

    Martin W Licht. On lipschitz partitions of unity and the assouad--nagata dimension. Topology and its Applications , 348:108845, 2024

  22. [22]

    Nagata dimension, quasisymmetric embeddings, and lipschitz extensions

    Urs Lang and Thilo Schlichenmaier. Nagata dimension, quasisymmetric embeddings, and lipschitz extensions. International Mathematics Research Notices , 2005(58):3625--3655, 2005

  23. [23]

    Lasota and James A

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

  24. [24]

    Spectral properties of dynamical systems, model reduction and decompositions

    Igor Mezi \'c . Spectral properties of dynamical systems, model reduction and decompositions. Nonlinear Dynamics , 41(1):309--325, 2005

  25. [25]

    Method of orthogonal projections and approximation of the spectrum of a bounded operator

    Andrzej Pokrzywa. Method of orthogonal projections and approximation of the spectrum of a bounded operator. Studia Math. , 65, 1979

  26. [26]

    Variance representations and convergence rates for data-driven approximations of koopman operators

    Friedrich M Philipp, Manuel Schaller, Septimus Boshoff, Sebastian Peitz, Feliks N \"u ske, and Karl Worthmann. Variance representations and convergence rates for data-driven approximations of koopman operators. arXiv preprint arXiv:2402.02494 , 2024

  27. [27]

    Spectral analysis of nonlinear flows

    Clarence W Rowley, Igor Mezi \'c , Shervin Bagheri, Philipp Schlatter, and Dan S Henningson. Spectral analysis of nonlinear flows. Journal of fluid mechanics , 641:115--127, 2009

  28. [28]

    Wold decomposition in banach spaces

    Alexander Vladimirovich Romanov. Wold decomposition in banach spaces. Mathematical Notes , 82(5):806--815, 2007

  29. [29]

    Variational analysis , volume 317

    R Tyrrell Rockafellar and Roger J-B Wets. Variational analysis , volume 317. Springer Science & Business Media, 2009

  30. [30]

    Dynamic mode decomposition of numerical and experimental data

    Peter J Schmid. Dynamic mode decomposition of numerical and experimental data. Journal of fluid mechanics , 656:5--28, 2010

  31. [31]

    Spectra and pseudospectra: the behavior of nonnormal matrices and operators

    Lloyd N Trefethen and Mark Embree. Spectra and pseudospectra: the behavior of nonnormal matrices and operators. 2020

  32. [32]

    Lipschitz algebras

    Nik Weaver. Lipschitz algebras . World Scientific Publishing Co., Inc., River Edge, NJ, 1999

  33. [33]

    Banach spaces for analysts

    Przemyslaw Wojtaszczyk. Banach spaces for analysts . Number 25. Cambridge University Press, 1996