Differential processes generated by two interpolators

J.M.F. Castillo; M. Gonz\'alez; V. Ferenczi; W.H.G. Correa

REVIEW 3 major objections 5 minor 20 references

Differential processes generated by two interpolators

T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read A single abstract commutator theorem unifies all known interpolation commutator estimates.

desk verdict A genuinely unifying framework for commutator theorems in interpolation theory; the abstract core is sound, though the transfer to stability and singularity results leans heavily on the authors' earlier work. read the letter →

arxiv 1908.09228 v1 pith:BQM5ZCAL submitted 2019-08-24 math.FA

J.M.F. Castillo , W.H.G. Correa , V. Ferenczi , M. González This is my paper · ORCID

classification math.FA MSC 46B7046E3046M18

keywords interpolationtheoryinterpolatorscommutatortheoremsdifferentialmethodstwistedsumsderivedspacestranslationoperatorsstabilityofexactsequences

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 studies pairs of interpolation methods, called interpolators, and the differential operators they generate. Its central claim is that for any two interpolators $\Psi$ and $\Phi$, the commutator of an arbitrary operator $\tau$ on the scale with the differential $\Omega_{\Psi,\Phi}$ is bounded between two canonically defined spaces, with one explicit estimate. The same theorem, the authors argue, yields every known commutator estimate in interpolation theory: for differential methods, for compatible interpolators, and for translation operators. The construction is symmetric in the two interpolators, and that symmetry lets stability and singularity results previously known for the complex method be carried over to general differential methods. A reader should care because a collection of ad hoc estimates is replaced by a single mechanism.

What carries the argument

The engine of the paper is the derivation $\Omega_{\Psi,\Phi}=\Psi\circ B_\Phi$, where $\Phi$ is an interpolator, $\Psi$ is a second interpolator on the same space $H$, and $B_\Phi$ is a bounded homogeneous selection map lifting each point of $X_\Phi=\Phi(H)$ back into $H$. Associated to the pair is the derived space $d\Omega_{\Psi,\Phi}=\Psi(\ker\Phi)\oplus_\Omega X_\Phi$, whose elements $(w,x)$ satisfy $w-\Omega_{\Psi,\Phi}x\in\Psi(\ker\Phi)$, together with the exact sequences (3) and (5). The argument is carried by the symmetry identities $\mathrm{Dom}(\Omega_{\Psi,\Phi})=\Phi(\ker\Psi)$ and $\mathrm{Ran}(\Omega_{\Psi,\Phi})=X_\Psi$, the bounded-splitting equivalences of Theorem 4.1, and the diagram chase in Theorem 7.1. The absence of compatibility or categorical assumptions is what allows the same pair machinery to cover compatible and non-compatible interpolators alike.

What would settle it

Take any pair of interpolators and an operator $\tau$ on the scale, and check whether $[\tau,\Omega_{\Psi,\Phi}]$ sends $\mathrm{Ran}(\Omega_{\Phi,\Psi})=X_\Phi$ into $\mathrm{Dom}(\Omega_{\Phi,\Psi})=\Phi(\ker\Psi)$ with the claimed norm bound; a single violation refutes the theorem's claimed generality. A concrete test case is a weighted function space with weights $w_0,w_1$ and a pointwise multiplier $\tau$ bounded on both weighted spaces but not on the interpolated space; if the predicted estimate fails there, the abstract theorem is wrong. Alternatively, in the annulus construction, compute the norm of division by $z-s$ on $J(X,B)$ for a concrete couple such as $\ell_p$: an unbounded division operator would falsify the bicontinuity claim of Proposition 8.6.

Watch

Extended reading notes

Core claim

The paper's central discovery is Theorem 7.1, the abstract commutator theorem. Given an interpolation couple, a pair of interpolators $(\Psi,\Phi)$ on the space $H$ of functions, and any operator $\tau$ acting on the scale, the commutator $[\tau,\Omega_{\Psi,\Phi}]=\tau\Omega_{\Psi,\Phi}-\Omega_{\Psi,\Phi}\tau$ is bounded from $\mathrm{Ran}(\Omega_{\Phi,\Psi})=X_\Phi$ into $\mathrm{Dom}(\Omega_{\Phi,\Psi})=\Phi(\ker\Psi)$, with norm at most $\max\{\|\tau:\Psi(\ker\Phi)\to\Psi(\ker\Phi)\|,\|\tau:X_\Phi\to X_\Phi\|,2\|T\|\|B_\Phi\|\}$. Equivalently, $\tau$ lifts to an operator on the derived space $X_{\Psi,\Phi}$, so the exact sequences commute. The proof is a diagram chase using the identities $\mathrm{Dom}(\Omega_{\Psi,\Phi})=\Phi(\ker\Psi)$ and $\mathrm{Ran}(\Omega_{\Psi,\Phi})=X_\Psi$ from Proposition 3.6. The authors' claim is that every earlier commutator theorem, in the differential-methods setting, the compatible-interpolator setting, and the translation-operator setting, is a specialization of this one result.

Load-bearing premise

For the core commutator theorem, everything rests on being able to choose a bounded lifting $B_\Phi$ that sends each interpolated value back to a function in the working space $H$; for the stability conclusions, one must also be able to divide functions by $z-s$ in the annulus space without losing control of the norm.

Editorial extensions

If this is right

Every commutator estimate for differential methods follows from Theorem 7.1 by taking $(\Psi,\Phi)$ to be the pair of evaluations attached to the annulus construction.
For compatible pairs, the theorem reduces to the standard form: $[\tau,\Omega_{\Psi,\Phi}]$ is bounded from $X_\Phi$ to $X_\Phi$, recovering the familiar estimate.
For translation operators $R_{\theta,\nu}=\Phi_\theta B_{\Phi_\nu}$, the commutator $[\tau,R_{\theta,\nu}]$ is bounded from $X_\nu$ into $\Phi_\theta(\ker\Phi_\nu)$, with a bound involving the gap between the two kernels.
Continuous families of interpolators preserve splitting and isomorphism of the derived spaces for nearby parameters, and bicontinuous pairs preserve the joint exact sequence.
The singularity and total-incomparability criteria previously proved for the complex method transfer to general differential methods.

Reading between the lines

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

If Theorem 7.1 is correct, future interpolation methods need only be checked for a bounded selection map and an invariance condition on the function space; all commutator estimates then come for free.
The symmetry between $\Omega_{\Psi,\Phi}$ and $\Omega_{\Phi,\Psi}$ suggests a duality principle for twisted sums: one underlying space carries two exact sequences with the roles of domain and range exchanged, so a centralizer bound on one side may automatically yield a bound on the other.
The gap-based bound for translation operators suggests a quantitative stability statement: if the kernels of two evaluations are close in gap, then every scale operator has small commutator with the translation, a feature that could be tested in weighted $\ell_p$ scales.
The paper leaves open whether triviality of $\Omega_{\Psi,\Phi}$ forces triviality of $\Omega_{\Phi,\Psi}$ (Problem 4.3); the symmetry established here makes a positive answer plausible but not proven.

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.

Desk Editor's Note

read the letter

This paper deserves a serious referee. It does what the best abstract interpolation papers do: it finds the single diagram that makes a scattered literature look like one theorem. The universal commutator theorem (7.1) is the real thing; the diagram chase checks out, and the symmetry results (Prop 3.6) and bounded splitting equivalence (Thm 4.1) are genuinely new organizing facts. Working without compatibility assumptions is exactly what lets translation operators and symmetric commutators fall out. The examples in Sections 5 and 7 are useful, and the weighted Kothe and Lorentz space computations show the framework is not empty.

The softest part is not Section 7 but the stability and singularity transfer in Section 8. The identification of the Cwikel–Kalton–Milman–Rochberg annulus evaluations with delta_s and delta'_s relies on [14, Lemma 3.11] for division by z-s, and on the authors' own previous work for the singularity criteria. If those imports wobble, the auxiliary conclusions wobble, while the abstract theorems stand. The reader flags a non-direct implication in Theorem 4.1: 'Clearly (3) implies (4)' is indeed not direct as printed; the intended route is via Proposition 3.6, so I'd call it a small presentational gap rather than a mathematical one. Proposition 7.3 has an evident typo in the proof (the gap estimate should put the gap on the pair of kernels, not the same kernel twice) and the display has a mangled H quotient; again fixable. There is also a healthy amount of self-citation, but in context the cited papers are the source of the singularity criteria and weighted-space formulas, so it is not a red flag.

Who this is for: people who work on interpolation theory, twisted sums, and commutators. It is an important contribution to Banach space interpolation theory. The central theorem holds up. I would accept it for peer review and engage with it myself. I'd ask the authors to fix the presentation issues and to make the dependence on imported results explicit in Section 8.

Referee Report

3 major / 5 minor

Summary. The paper develops an abstract framework for interpolation methods without compatibility or categorical assumptions. For a pair of interpolators (Ψ,Φ) on a function space H, it defines the derivation ΩΨ,Φ = ΨBΦ, the derived quasi-Banach space dΩΨ,Φ, and the associated exact sequences. Proposition 3.6 identifies the domain of the derivation with Φ(kerΨ) and its range with XΨ. Theorem 4.1 gives algebraic and topological equivalences for bounded splitting of the induced sequences. The main result, Theorem 7.1, is a 'universal' commutator theorem: for every operator τ acting on the scale, [τ,ΩΨ,Φ] maps Ran(ΩΦ,Ψ) = XΦ into Dom(ΩΦ,Ψ) = Ψ(kerΦ) with the displayed bound, obtained by a diagram chase. The remainder of the paper applies this to CKMR differential methods, translation operators, weighted Köthe spaces, Lorentz and Orlicz spaces, and derives stability and singularity results in terms of kernel gaps.

Significance. If Theorem 7.1 is correct, it is a genuinely unifying result: it recovers and extends the commutator theorems of Cwikel-Kalton-Milman-Rochberg and Carro-Cerdà-Soria from a single exact-sequence argument, and it reveals the symmetric role of the pair (Ψ,Φ) and (Φ,Ψ). The paper is self-contained up to standard background on exact sequences, and the main diagram chase is valid; I found no flaw in the central commutator theorem. The framework's economy is a strength, as are the explicit identifications of Dom and Ran in Sections 3 and 5. The main qualification is in Section 8, where the transfer to the CKMR annulus methods is less complete, so the paper's stability claims are not yet fully supported as printed.

major comments (3)

[Section 4, Theorem 4.1] The implication '(3) ⇒ (4)' is not immediate as printed. From (3), XΨ = Ψ(kerΦ), one first has to prove H = kerΦ + kerΨ: for f ∈ H choose g ∈ kerΦ with Ψf = Ψg; then f − g ∈ kerΨ. Then (2) XΦ = Φ(kerΨ) follows, and Proposition 3.6(1) gives (4). Since (2)⇔(4) is used in the rest of the proof, the one-line 'Clearly' leaves a gap in a load-bearing equivalence. The repair is short, but it should be included.
[Section 8, Proposition 8.6] The proof of bicontinuity is not complete. The argument uses division by (z−s) to define g and h, appealing to [14, Lemma 3.11], but it neither states the hypotheses of that lemma nor proves the needed boundedness of multiplication by (z−t) and by (z−t)^2 on J(X,B); these boundedness properties are exactly what is needed to conclude that (z−t)g(z) ∈ kerΦ_t and (z−t)^2g(z) ∈ kerΨ_t ∩ kerΦ_t with controlled norms. In addition the displayed inequality ‖f(z)−(z−t)^2g(z)‖ = |(z−s)^2−(z−t)^2|‖g‖ ≤ (|s|^2−|t|^2+2z|t−s|)C^2‖f‖ is not a valid estimate: it mixes complex scalars with vectors, depends on z, and is not a uniform bound on the annulus. A correct argument would bound sup_{z∈A}|(2z−s−t)(t−s)| by a constant times |t−s|. As printed, Proposition 8.6 does not establish the continuity and bicontinuity on which Propositions 8.4 and 8.5 and the stability transfer depend.
[Section 7, Proposition 7.3] The displayed estimate contains an apparent typo and a missing step. The factor g(kerΦθ, kerΦθ) should be the gap between kerΦν and kerΦθ (or an equivalent quantity), and the inequality dist(TB_{Φν}(x)−B_{Φν}τx, kerΦθ) ≤ ‖TB_{Φν}(x)−B_{Φν}τx‖_H g(kerΦν, kerΦθ) requires an explanation of the relationship between the quotient norm on Φθ(H) and the gap between the kernels. Without this justification, the stated bound on [τ,Rθ,ν] does not follow from Theorem 7.1 in the form printed.

minor comments (5)

[Section 3, proof of Proposition 3.4] In the proof of surjectivity of Q, the statement 'w−ΩΨ,Φx ∈ XΦ' should read 'w−ΩΨ,Φx ∈ Ψ(kerΦ)'.
[Section 5.4, Claim 3] In the proof of Claim 3, the text '‖ = max{...}' is a typographical corruption of the norm equality and should be corrected.
[Section 8.2, proof of Proposition 8.9] In the proof of item (1), 'C^{-1}piM' should read 'C^{-1}πM'.
[Section 7.3 and Problem 8.10] The paper defers proofs to unpublished references [5] and [7]; if these are not available to the reader, the corresponding claims are not self-contained. The authors should either include the deferred arguments or mark the reliance explicitly.
[Section 3 and Section 7] The spaces XΦ,Ψ and XΨ,Φ are both used with the coordinate swap left implicit; a sentence in Section 3 explaining the isomorphism would reduce confusion in the statement and proof of Theorem 7.1.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the abstract commutator theorem is derived from definitions and diagram chases, not from the commutator estimates it unifies.

full rationale

The central claim, Theorem 7.1, is proved directly from the definitions of the interpolators, the derivation Omega_{Psi,Phi} = Psi B_Phi, and the exact sequence structure of X_{Psi,Phi}; the boundedness of [tau, Omega_{Psi,Phi}] is computed from the interpolated operator T and the selection map B_Phi, not assumed. The domain and range identities in Proposition 3.6 are proved from the definitions of Dom and Ran, and Proposition 3.2 gives the equivalence between the two exact-sequence presentations. The known commutator estimates for differential methods, compatible interpolators, and translation operators appear as consequences of the abstract theorem after explicit identifications in Sections 5 and 7, so the derivation does not presuppose those estimates. The paper does cite prior work by the same authors, especially [8, 9, 12] for singularity criteria and [9] for weighted-space interpolation formulas, and it defers some determinations to [5] and [7]; these are normal uses of prior published results with independent proofs and are explicitly acknowledged limitations, not circular reductions. No fitted parameter is renamed as a prediction, no uniqueness theorem is imported from the authors' prior work, and no ansatz is smuggled in via citation. The abstract framework and its central theorem are self-contained, and the generalization claims rest on the quoted lemmas rather than on the conclusions being derived.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

No free parameters fitted to data appear in this mathematical framework. The constants epsilon, C, and c in the proofs are universal norm constants, not adjustable parameters. No new empirical entities are postulated; the derived spaces, domains, and ranges are defined objects backed by theorems rather than entities introduced to absorb contradictions.

assumptions (6)

domain assumption For every interpolator Phi there exists a homogeneous selection B_Phi : X_Phi -> H with Phi B_Phi x = x and norm bound ||B_Phi x||_H <= (1+epsilon)||x||_Phi.
Introduced in Section 3 before equation (2); it is standard for quotient maps, but the definition of Omega and the norm estimates in Theorem 7.1 depend on the boundedness of B_Phi.
domain assumption The abstract interpolation method axioms: H is a Banach space of functions on a metric space D into Sigma, every scale operator t induces a bounded T on H, and each interpolator Phi : H -> Sigma satisfies t composed with Phi equals Phi composed with T.
Section 3 defines the setting; without these commutation relations, diagram (1) and Theorem 7.1 fail.
standard math Kato's perturbation theorems on gap and minimum gap, including the openness of complementation with respect to the gap, are used as stated in [19, Chapter IV].
Proposition 2.1 is proved using [19, IV Theorem 4.2 and IV Lemma 4.4], and these results support the stability arguments in Section 8.
standard math For the Cwikel-Kalton-Milman-Rochberg differential method, the space J(X,B) is identified with analytic functions on the annulus A, and division by z-s is a bounded operation with constant independent of f, as in [14, Lemma 3.11].
Used in Proposition 8.6 to prove continuity and bicontinuity of the interpolator families; if division by z-s were not bounded, the stability arguments would not transfer.
domain assumption The singularity estimates [8, Lemma 4.8] and [12, Lemma 2.11] transfer from the complex strip method to the CKMR annulus setting, and Proposition 8.8 is quoted from [8,12].
Section 8.2 asserts the transfer without a full derivation; this is the basis for the singularity criteria in Proposition 8.8.
standard math In Example 5.4, the identity X_theta = X(w_theta) for weighted Koethe spaces is imported from [9, Proposition 4.1].
Used to identify the derivation Omega with multiplication by log(w1/w0) and to compute the symmetric derivation.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Differential processes generated by two interpolators." pith.science (2026). https://pith.science/paper/BQM5ZCAL

@misc{pith2026190809228,
  author       = {Pith},
  title        = {Pith review of: Differential processes generated by two interpolators},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BQM5ZCAL}},
  note         = {Machine review of arXiv:1908.09228}
}

read the original abstract

We study couples of interpolators, the differentials they generate and their associated commutator theorems. An essential part of our analysis is the study of the intrinsic symmetries of the process. Since we work without any compatibility or categorical assumption, our results are flexible enough to generalize most known results for commutators or translation operators, in particular those of Cwikel, Kalton, Milman, Rochberg \cite{ckmr} for differential methods and those of Carro, Cerd\`a and Soria \cite{caceso} for compatible interpolators. We also generalize stability and singularity results in \cite{cfg,ccfg,correa} from the complex method to general differential methods and obtain new incomparability results.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

20 extracted references · 20 canonical work pages

[1]

Bergh, J

J. Bergh, J. L¨ ofstr¨ om,Interpolation spaces. An introduction. Springer, 1976

work page 1976
[2]

Cabello, Nonlinear centralizers with values in L0, Nonlinear Analysis - TMA 88 (2013) 42–50

F. Cabello, Nonlinear centralizers with values in L0, Nonlinear Analysis - TMA 88 (2013) 42–50

work page 2013
[3]

Cabello, Factorization in Lorentz spaces, with an application to cen tralizers, J

F. Cabello, Factorization in Lorentz spaces, with an application to cen tralizers, J. Math. Anal. Appl. 446 (2017) 1372–1392

work page 2017
[4]

Cabello, Nonlinear centralizers in homology , Math

F. Cabello, Nonlinear centralizers in homology , Math. Ann. 358 (2014) 779 – 798

work page 2014
[5]

Cabello, W

F. Cabello, W. Corrˆ ea Quasinormed groups and Fenchel-Orlicz spaces , to appear

work page
[6]

M. J. Carro, J. Cerda, F. Soria, Commutators in interpolation theory, Arkiv. Mat. 33 (1995) 199-216

work page 1995
[7]

Higher order derivatives of analytic families of Banach spaces

J.M.F. Castillo, W.H.G. Correa, F. Cabello, Higher order derivatives of analytic families of Banach spa ces, Arxiv submit/1906.06677, 2019

work page Pith review arXiv 1906
[8]

Castillo, V

J.M.F. Castillo, V. Ferenczi, M. Gonz´ alez, Singular twisted sums generated by complex interpolation , Trans. Amer. Math. Soc. 369 (2017) 4671–4708. DIFFERENTIAL PROCESSES GENERATED BY TWO INTERPOLATORS 21

work page 2017

Show all 20 references

[9]

Castillo, W.H.G

J.M.F. Castillo, W.H.G. Correa, V. Ferenczi, M. Gonz´ alez,On the stability of the diﬀerential process generated by complex interpolation , Arxiv submit/1712.09647, 2019

2019 arXiv
[10]

Castillo, M

J.M.F. Castillo, M. Gonz´ alez, Three-space problems in Banach space theory, Lecture Notes in Math. 1667. Springer, 1997

1997
[11]

Cerd` a, A note on commutator estimates for interpolation methods , Math

J. Cerd` a, A note on commutator estimates for interpolation methods , Math. Nachr. 280 (2007) 1014–1021

2007
[12]

Corrˆ ea,Type, cotype and twisted sums induced by complex interpolat ion, J

W.H.G. Corrˆ ea,Type, cotype and twisted sums induced by complex interpolat ion, J. Funct. Anal. 274 (2018) 659–958

2018
[13]

Analysis at Urbana II

M. Cwikel, B. Jawerth, M. Milman, R. Rochberg, Diﬀerential estimates and commutators in interpolation theory. In “Analysis at Urbana II”, London Math. Soc. Lecture Note Serie s 138, (E.R. Berkson, N.T. Peck, and J. Uhl, Eds.), pp. 170–220, Cambridge Univ. Press, Cambridge, 1989

1989
[14]

Cwikel, N.J

M. Cwikel, N.J. Kalton, M. Milman, R. Rochberg, A uniﬁed theory of Commutator Estimates for a class of interpolation methods, Adv. in Math. 169 (2002) 241-312

2002
[15]

Androulakis and C.D

G. Androulakis and C.D. Cazaku and N.J. Kalton, Twisted sums, Fenchel-Orlicz spaces and property (M). Houston J. Math. 24 (1998), no. 1, 105–126

1998
[16]

Gustavsson and J

J. Gustavsson and J. Peetre, Interpolation of Orlicz spaces. Studia Math. 60 (1977) 33–59

1977
[17]

Kalton, S

N.J. Kalton, S. Montgomery-Smith, Interpolation of Banach spaces, Chapter 36 in Handbook of the Geometry of Banach spaces vol. 2, (W.B. Johnson and J. Lindenstrauss eds.) , pp. 1131–1175, North-Holland 2003

2003
[18]

Kalton, N.T

N.J. Kalton, N.T. Peck, Twisted sums of sequence spaces and the three space problem, Trans. Amer. Math. Soc. 255 (1979) 1–30

1979
[19]

Kato, Perturbation theory for linear operators, Corrected printing of the 2nd

T. Kato, Perturbation theory for linear operators, Corrected printing of the 2nd. ed. Springer, 1980

1980
[20]

Krugljak, M

N. Krugljak, M. Milman, A distance between orbits that controls commutator estimat es and invertibility of operators, Adv. Math. 182 (2004) 78-123. Instituto de Matem ´aticas, Universidad de Extremadura, A venida de Elvas s/n, 0 6011 Badajoz, Spain. E-mail address : castillo@u...

2004

Pith tools

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

[1] [1]

Bergh, J

J. Bergh, J. L¨ ofstr¨ om,Interpolation spaces. An introduction. Springer, 1976

work page 1976

[2] [2]

Cabello, Nonlinear centralizers with values in L0, Nonlinear Analysis - TMA 88 (2013) 42–50

F. Cabello, Nonlinear centralizers with values in L0, Nonlinear Analysis - TMA 88 (2013) 42–50

work page 2013

[3] [3]

Cabello, Factorization in Lorentz spaces, with an application to cen tralizers, J

F. Cabello, Factorization in Lorentz spaces, with an application to cen tralizers, J. Math. Anal. Appl. 446 (2017) 1372–1392

work page 2017

[4] [4]

Cabello, Nonlinear centralizers in homology , Math

F. Cabello, Nonlinear centralizers in homology , Math. Ann. 358 (2014) 779 – 798

work page 2014

[5] [5]

Cabello, W

F. Cabello, W. Corrˆ ea Quasinormed groups and Fenchel-Orlicz spaces , to appear

work page

[6] [6]

M. J. Carro, J. Cerda, F. Soria, Commutators in interpolation theory, Arkiv. Mat. 33 (1995) 199-216

work page 1995

[7] [7]

Higher order derivatives of analytic families of Banach spaces

J.M.F. Castillo, W.H.G. Correa, F. Cabello, Higher order derivatives of analytic families of Banach spa ces, Arxiv submit/1906.06677, 2019

work page Pith review arXiv 1906

[8] [8]

Castillo, V

J.M.F. Castillo, V. Ferenczi, M. Gonz´ alez, Singular twisted sums generated by complex interpolation , Trans. Amer. Math. Soc. 369 (2017) 4671–4708. DIFFERENTIAL PROCESSES GENERATED BY TWO INTERPOLATORS 21

work page 2017