Pith. sign in

REVIEW 2 major objections 4 minor 15 references

Korovkin-type results on convergence of sequences of positive linear maps on function spaces

T0 review · 2 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Pith's one-line read This paper proves generalized Korovkin-type theorems: pointwise or uniform convergence of positive linear maps on a test function space forces convergence on the whole space, even when the limit is only assumed to be an isometry.

desk verdict A genuine extension of Korovkin-type results, with a solid real-scalar proof and a genuine gap in the complex case: the proof calls on an unstated classification theorem whose hypotheses are not checked. read the letter →

arxiv 1908.03027 v1 pith:DLFHFEIQ submitted 2019-08-08 math.FA

classification math.FA MSC 41A3646E15
keywords KorovkintheorempositivelinearmapsChoquetboundaryfunctionspaceisometrypointwiseconvergenceuniformapproximationtheory
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

Approximation theory's classical Korovkin theorem says that positive linear operators that converge on the handful of test functions $1,x,x^2$ converge on every continuous function. This paper proves a broad extension: a sequence of positive linear maps from a function space $M$ to $C(Y)$ that converges pointwise on a smaller test space $S$ to an isometry $T_\infty$—which need not be assumed linear—also converges pointwise on the Choquet boundary of $T_\infty(S)$ for every $f\in M$. With two mild extra conditions, $\operatorname{Ch}(N)\subseteq\operatorname{Ch}(T_\infty(S))$ and boundedness of $\{T_n1\}$, the convergence extends to all of $Y$; the uniform version holds on compact subsets of the boundary, and globally when the boundary is compact. The authors establish this by first showing the limit isometry is actually linear and has the form $T_\infty f=f\circ\varphi$, then using positivity and peak test functions to push convergence from $S$ to all of $M$.

What carries the argument

The engine is the Choquet boundary $\operatorname{Ch}(S)$ together with the peak-function criterion of Lemma 2.2: a point $x_0$ lies in $\operatorname{Ch}(S)$ exactly when one can find functions in $S$ that are small at $x_0$ and large outside any neighbourhood of it. The proof uses this to build the test functions $f_{y'}$ that sandwich $f$ through the tensor-product positivity inequality $|(T_n\otimes_{T_n1}T_\infty)(F)| \le (T_n\otimes_{T_n1}T_\infty)(1\otimes\epsilon+\|F\|\,\operatorname{Re} F_{y'})$. Linearity of $T_\infty$ is forced via the Mazur–Ulam theorem and a classification of real-linear isometries; a boundary-measure representation then carries convergence from $\operatorname{Ch}(T_\infty(S))$ to all of $Y$, and the Arzelà–Ascoli compactness criterion upgrades pointwise to uniform convergence on compact subsets.

What would settle it

One concrete test: construct a real-linear isometry $T_\infty$ between subspaces of $C(X)$ with $T_\infty 1 \ge 0$ but $T_\infty i \ne iT_\infty 1$, together with positive linear maps $T_n$ that converge to it on the test space $S$. If the hypotheses of Theorem 3.1 hold yet convergence fails outside $S$, the linearity step—and with it the appeal to the classification of real-linear isometries—is the point to inspect. A simpler check of that step is to verify whether the cited classification theorem actually applies to arbitrary self-conjugate unital subspaces $M$ used here.

Watch

Extended reading notes

Core claim

The central discovery is Theorem 3.1: under the stated hypotheses, the limit isometry $T_\infty$ is forced to be a linear isometry of the form $T_\infty f = f\circ\varphi$ on the Choquet boundary of $T_\infty(M)$, and once that representation is available, a positivity sandwich forces convergence from $S$ to all of $M$. Part (a) states $T_nf\to T_\infty f$ on $\operatorname{Ch}(T_\infty(S))$ for every $f\in M$; part (b) upgrades this to all of $Y$ when $\operatorname{Ch}(N)\subseteq\operatorname{Ch}(T_\infty(S))$ and $\{T_n1\}$ is bounded. Theorem 3.2 is the uniform analogue: uniform convergence on $S$ gives uniform convergence on compact subsets of $\operatorname{Ch}(T_\infty(S))$, and global uniform convergence when $\operatorname{Ch}(T_\infty(S))$ or $\operatorname{Ch}(N)$ is compact. The examples show the classical Korovkin theorem, its complex variants on the torus and disk, and a smooth multivariate version all follow from the same mechanism.

Load-bearing premise

The load-bearing premise is that the external classification of real-linear isometries between subspaces of continuous functions applies to the self-conjugate unital subspaces M allowed in the theorem; the paper invokes it without restating its hypotheses, and if it does not apply, T∞ need not be linear and the representation T∞f = f∘φ—the engine of Steps 2 and 3—fails.

Editorial extensions

If this is right

  • The classical theorem for $C[0,1]$ follows as the case $M=S=\operatorname{Span}\{1,x,x^2\}$ with $T_n$ converging to the identity, so the quadratic test functions alone force convergence for every continuous function.
  • Complex Korovkin-type statements on the torus and disk are recovered: convergence of $T_n1\to1$ and $T_nz\to z$ forces convergence on $C(\mathbb{T})$, and adding $T_n|z|^2\to|z|^2$ does the same on $C(\mathbb{D})$.
  • For smooth functions $D_K$ on a compact subset of $\mathbb{R}^p$, convergence on $1$, the coordinate projections $P_k$, and $\sum P_k^2$ forces convergence for every $C^\infty$ function on $K$.
  • Pointwise convergence on $S$ propagates to all of $Y$ when $\operatorname{Ch}(N)\subseteq\operatorname{Ch}(T_\infty(S))$ and $\{T_n1\}$ is bounded; uniform convergence on $S$ propagates uniformly on compact subsets of $\operatorname{Ch}(T_\infty(S))$, and globally when the relevant Choquet boundary is compact.
  • The proofs work for nets as well as sequences, so the results are sequential only in presentation.

Reading between the lines

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

  • The paper does not pursue rates, but the proof's explicit $\epsilon$ control suggests that a quantitative version—uniform modulus of convergence on compact subsets in terms of the test-space convergence—can be extracted from the inequalities in Theorem 3.2.
  • Because Step 2 is the only place the isometry is shown to be linear, a promising stress test is to search for real-linear isometries of subspaces that satisfy $T_\infty 1\ge0$ but are not complex-linear; if one is also a pointwise limit of positive linear maps, it would mark exactly where the classification hypothesis bites.
  • The transfer mechanism via tensor-product positivity is not tied to $C(X)$ in any essential way, so analogues should hold for positive maps on other Banach function spaces that admit a Choquet-type boundary, such as spaces of differentiable functions or Lipschitz spaces.
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

2 major / 4 minor

Summary. The paper studies Korovkin-type convergence for sequences of positive linear maps T_n from a self-conjugate subspace M of C(X) into C(Y). The limit map T∞ is assumed only to be an isometry from M onto its image, not a linear operator. Theorem 3.1(a) claims pointwise convergence of T_n f to T∞f on the Choquet boundary of T∞(S) whenever this convergence holds on the test space S; part (b) uses a boundary inclusion plus boundedness of {T_n 1} to extend convergence to all of Y. Theorem 3.2 gives the uniform version, Corollary 3.4 recovers earlier results of Hachiro and Okayasu, and the examples include classical Korovkin-type statements.

Significance. If correct, the paper broadens the usual Korovkin framework in two useful ways: the approximating maps need only be positive rather than unital contractions, and the limit map need not be assumed linear. The use of the quotient by N and of extreme-boundary measures in Theorem 3.1(b) is a clever device, and the examples illustrate the intended scope. The contribution is potentially interesting, but the manuscript as written cannot be certified: the proof of the decisive linearity step depends on an unverified application of an external classification theorem, and one later passage in Step 3 uses the Choquet boundary in a way that is not justified.

major comments (2)
  1. [Theorem 3.1, Step 2] The proof of the linearity of T∞ is incomplete as written. After Mazur-Ulam gives real-linearity, the manuscript invokes [9, Theorem 2.3] without stating its hypotheses and without verifying that the pair (M, T∞(M)) satisfies them. In particular, if that theorem requires the codomain to be a function space on Y, or a strongly separating subspace, then the conclusion cannot be applied to T∞:M→T∞(M): an isometry of the form T∞f = f∘π with π:Y→X continuous, surjective and non-injective has range isometric to M but this range does not separate the fibres of π. Since Step 2 is the only place where T∞1 = 1 and complex-linearity of T∞ are established, and since Step 3 and part (b) both use that linearity, the central claim of Theorem 3.1 is not justified unless the hypotheses of [9, Theorem 2.3] are stated and checked. Theorem 3.2 inherits the same dependence.
  2. [Theorem 3.1, Step 3] The passage 'it is observed that the above relation holds for all z, z′ ∈ Y' is not justified. The displayed inequality before this sentence is established only for z′ ∈ Ch(T∞(M)); the Choquet boundary property of T∞(M) does not by itself imply that a real-valued function in T∞(M) is nonnegative on all of Y whenever it is nonnegative on Ch(T∞(M)). Moreover, the final estimate is later applied at y′ ∈ Ch(T∞(S)), which need not lie in Ch(T∞(M)): for a subspace inclusion the Choquet boundary of the smaller space can be strictly larger. Thus the proof needs an explicit argument for this step, or the final estimate needs to be derived directly with z′ = y′.
minor comments (4)
  1. [Theorem 3.1, Step 2] The displayed representation of T∞ writes f∘φ in both alternatives on K and on its complement; one of the two cases should be the conjugate function −̅{f∘φ} or an equivalent expression.
  2. [Theorem 3.2, Proof of Claim] In the displayed equicontinuity estimate, the term η T∞ f_{y′}(y) appears where a real part such as η Re T∞ f_{y′}(y) is expected, and the symbol η is defined only after it is used; please reorder and correct the displayed inequality.
  3. [Theorem 3.1(b)] The notation R̂{Ch}(T∞(S)) is used without definition; please define it as the image of Ch(T∞(S)) under the quotient map and explain the inclusion Ch(R̂N) ⊆ R̂{Ch}(T∞(S)).
  4. [Example 4.3] The peak function for z0 ∈ T is not h(z) = (z + z0)/2, since h(z0) = z0; it should be (1 + ̅z0 z)/2, or an equivalent rotation, in order to satisfy h(z0) = 1 and |h(z)| < 1 for z ≠ z0.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the main theorem rests on external isometry-classification and prior Korovkin-type lemmas, with only a non-load-bearing self-citation to [3].

full rationale

The derivation does not assume its conclusion. Theorem 3.1 assumes pointwise convergence on S and derives it on Choquet boundaries, which is the standard structure of a Korovkin theorem rather than a circular reduction. Step 2's proof that T∞ is linear uses the Mazur-Ulam theorem [10] and the external classification [9, Theorem 2.3]; this is independent support because that classification concerns real-linear isometries and does not contain the target convergence statement. Step 3 and Corollary 3.4 invoke [7, Lemma 2.5] (or [3, Corollary 3.2]) to obtain the representation T∞f=f∘φ on Choquet boundaries; [7] is an external source, so the self-citation [3] is not load-bearing. Part (b) uses Choquet measure representations from [4] and [12] and follows the closing argument of [7, Theorem 3.3], again external. No fitted parameter is renamed as a prediction, and no uniqueness theorem is imported from the authors' own prior work. The only caveat, unverified hypotheses of [9, Theorem 2.3] in Step 2, is a correctness or rigor concern rather than evidence of circularity.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

The central theorems are proved from standard functional-analysis tools plus three external results about isometries on Choquet boundaries. The heaviest external load is the classification theorem [9, Theorem 2.3], which is cited without a statement of its hypotheses. No free parameters or invented entities appear.

assumptions (7)
  • standard math Mazur-Ulam theorem: a surjective isometry between normed spaces that fixes 0 is real-linear.
    Used in Theorem 3.1, Step 2, to pass from isometry to real-linearity after showing T∞0=0.
  • domain assumption [9, Theorem 2.3] classification of real-linear isometries between subspaces of continuous functions.
    External theorem invoked in Step 2 to get T∞1=1 and the representation T∞f = f∘φ or f∘conj∘φ on a clopen partition of Ch(T∞(M)). Its hypotheses are not reproduced in the paper.
  • domain assumption [7, Lemma 2.5] or [3, Corollary 3.2]: a linear isometry between function spaces induces a continuous surjection between Choquet boundaries with T∞f = f∘φ on the boundary.
    Used in Step 3 to transfer convergence from S to Ch(T∞(S)).
  • standard math Browder, Lemma 2.2: characterization of Choquet boundary points via peak-like functions.
    Used repeatedly to construct functions fy' satisfying Refy'≥0, Refy'≥1 off a neighborhood, and Refy' small at the chosen point.
  • standard math Arzela-Ascoli theorem.
    Used in Theorem 3.2 to upgrade pointwise convergence to uniform convergence on compact sets.
  • standard math Bishop-de Leeuw representation of points by measures on extreme boundaries, as presented in Phelps's lectures.
    Used in Theorem 3.1(b) to pass from convergence on Ch(T∞(S)) to all of Y via a quotient space.
  • domain assumption Standing structural hypotheses: S is a unital separating function space, M is a self-conjugate subspace, and Tn are positive linear maps.
    These are stated hypotheses of the main theorems. They are not proved, and the statements are not true without them.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Korovkin-type results on convergence of sequences of positive linear maps on function spaces." pith.science (2026). https://pith.science/paper/DLFHFEIQ

@misc{pith2026190803027,
  author       = {Pith},
  title        = {Pith review of: Korovkin-type results on convergence of sequences of positive linear maps on function spaces},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DLFHFEIQ}},
  note         = {Machine review of arXiv:1908.03027}
}
read the original abstract

In this paper we deal with the convergence of sequences of positive linear maps to a (not assumed to be linear) isometry on spaces of continuous functions. We obtain generalizations of known Korovkin-type results and provide several illustrative examples.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

15 extracted references · 15 canonical work pages

  1. [7]

    Hachiro and T

    T. Hachiro and T. Okayasu, Some theorems of Korovkin type , Studia Math. 155 (2003), no. 2, 131-143

  2. [9]

    Koshimizu, T

    H. Koshimizu, T. Miura, H. Takagi and S. E. Takahasi, Real-linear isometries between subspaces of continuous functions, J. Math. Anal. Appl. 413 (2014), 229-241

  3. [1]

    Altomare, Korovkin-type theorems and approximation by positive line ar operators, Surv

    F. Altomare, Korovkin-type theorems and approximation by positive line ar operators, Surv. Approx. Theory. 5 (2010), 92-164

  4. [2]

    Altomare and M

    F. Altomare and M. Campiti, Korovkin-Type Approximation Theory and its Applications , de Gruyter Studies in Mathematics, 17, W alter de Gruyter Co., Berlin, 1994

  5. [3]

    Araujo and J

    J. Araujo and J. J. Font, Linear isometries between subspaces of continuous functio ns, Trans. Amer. Math. Soc. 349 (1997), 413-428

  6. [4]

    Bishop and K

    E. Bishop and K. de Leeuw, The representation of linear functionals by measures on set s of extreme points , Ann. Inst. Fourier (Grenoble) 9 (1959), 305-331

  7. [5]

    Browder, Introduction to Function Algebras , W

    A. Browder, Introduction to Function Algebras , W. A. Benjamin, New York-Amsterdam, 1969

  8. [6]

    Donner, Extension of Positive Operators and Korovkin Theorems , Lecture Notes in Math

    K. Donner, Extension of Positive Operators and Korovkin Theorems , Lecture Notes in Math. 904, Springer- Verlag, Berlin, 1982

Show all 15 references
  1. [8]

    P. P. Korovkin, On convergence of linear positive operators in the space of c ontinuous functions , Doklady Akad. Nauk SSSR (N.S.) 90 (1953), 961-964 (in Russian)

  2. [10]

    Mazur and S

    S. Mazur and S. Ulam, Sur les transformations isom´ etriques d’espaces vectoriels norm´ es, C. R. Math. Acad. Sci. Paris 194 (1932), 946-948

  3. [11]

    E. N. Morozov, Convergence of a sequence of positive linear operators in th e space of continuous 2 π -periodic functions of two variables , Kalinin. Gos. Ped. Inst. Uchen. Zap. 26 (1958), 129-142 (in Russian)

  4. [12]

    R. R. Phelps, Lectures on Choquet’s Theorem, 2nd ed., Lecture Notes in Math. 1757, Springer, Berlin, 200 1

  5. [13]

    R. R. Phelps, The range of Tf for certain linear operators T , Proc. Amer. Math. Soc. 16 (1965), 381-382

  6. [14]

    Scheffold, Uber die punktweise konvergenz von operatoren in C(X), Rev

    E. Scheffold, Uber die punktweise konvergenz von operatoren in C(X), Rev. Acad. Ci. Zaragoza 28 (1973), 5-12

  7. [15]

    V. I. Volkov, Conditions for convergence of a sequence of positive linear operators in the space of continuous functions of two variables , Kalinin. Gos. Ped. Inst. Uchen. Zap. 26 (1958), 27-40 (in Russian). Department of Mathematics, K. N. Toosi University of Technolo gy, Teh...

Pith tools

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