REVIEW 3 major objections 6 minor 34 references
Ulam-Hyers stabilities of mild solutions of the fractional nonlinear abstract Cauchy problem
T0 review · 3 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read The paper proves that mild solutions of a fractional nonlinear abstract Cauchy problem are Ulam-Hyers stable under a contraction condition, with explicit stability constants on $[0,T]$ and $[0,\infty)$.
desk verdict Standard Ulam-Hyers stability theorems for a Hilfer abstract Cauchy problem, but the proofs rest on an ill-typed Lipschitz condition and dropped weight factors, so as written the contraction estimates don't hold. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the fixed-point operator $\Lambda(\xi)(t)=S_{\alpha,\beta}(t)\xi_0+\int_0^t T_\alpha(t-s)u(s)H(s,\xi(s))\,ds$ acting on the weighted continuous function space $C_{1-\gamma}(I,\Omega)$, whose norm is $\|\xi\|_{C_{1-\gamma}}=\sup_t\|t^{1-\gamma}\xi(t)\|$. The proof shows $\Lambda$ is a contraction either in the metric induced by this norm, for the constant-tolerance theorems, or in the $\varphi$-weighted metric $d_{\varphi,1-\gamma}$, for the Rassias-type theorems, and then invokes the Banach fixed point theorem. The stability estimate is obtained by applying the triangle inequality to the distance between an approximate solution and the fixed point, which reduces the approximate error to $c\varepsilon$ with $c=1/(1-\lambda)$. The kernel $T_\alpha(t)=t^{\alpha-1}G_\alpha(t)$ with $G_\alpha$ defined through the Mainardi-Wright function, together with the resolvent family $S_{\alpha,\beta}$, carries all the fractional-semigroup information.
What would settle it
Take the scalar model $A=0$, $u(t)=1$, $H(t,x)=ax$ with a fixed constant $a$. Compute the exact solution and compare the theorem's predicted Ulam-Hyers bound with the true weighted distance between an approximate solution, such as the exact solution plus a small bump of size $\varepsilon$, and the exact solution. A single instance with $\tilde\lambda<1$ whose weighted distance exceeds $\varepsilon/(1-\tilde\lambda)$ would refute the claim.
Extended reading notes
Core claim
The central claim is that the integral equation $\xi(t)=S_{\alpha,\beta}(t)\xi_0+\int_0^t T_\alpha(t-s)u(s)H(s,\xi(s))\,ds$ is stable in the Ulam-Hyers sense on $[0,T]$ whenever $\tilde\lambda=\delta T^{1-\gamma}\int_0^T e^{w(T-s)}|u(s)|\ell(s)\,ds<1$, and on $[0,\infty)$ whenever $\tilde\lambda_{\alpha,1-\gamma}=\sup_{t\ge0} t^{1-\gamma}\int_0^t \ell(s)|u(s)|\|T_\alpha(t-s)\|\,ds<1$. Here $S_{\alpha,\beta}$ is the $(\alpha,\beta)$-resolvent family generated by $A$ and $T_\alpha$ is built from the Mainardi-Wright function. The paper also claims the Ulam-Hyers-Rassias analogues with function-valued tolerances $G(t)$ or $\varphi(t)$, under the boundedness conditions (3.5) and (4.9) and the integral inequality $\int_0^t \varphi(s)\,ds\le K\varphi(t)$. In all four theorems the stability constant is explicit and has the form $1/(1-\text{contraction constant})$.
Load-bearing premise
The load-bearing premise is that the nonlinearity is Lipschitz; as written, the condition uses the function-space norm $\|x-y\|_{C_{1-\gamma}}$ for pointwise values, which requires an extra weight-handling step that the proofs do not supply before the contraction estimates follow.
Editorial extensions
If this is right
- On the finite interval, the paper implies that any function whose residual $t^{1-\gamma}(\xi-\Lambda\xi)$ is bounded by $\varepsilon$ lies within $\varepsilon/(1-\tilde\lambda)$ of a genuine mild solution, so the stability error is controlled by the data error with an explicit amplification factor.
- On the half-line, the uniform contraction condition $\tilde\lambda_{\alpha,1-\gamma}<1$ gives the same control uniformly in time, so stability does not deteriorate as $t\to\infty$.
- The parameter limits $\beta\to1$ and $\beta\to0$ recover the same stability theorems for the Caputo and Riemann-Liouville fractional derivatives, and $\alpha=1$ recovers the integer-order abstract Cauchy problem.
- The Rassias formulations imply that if the allowed error grows like a function $G(t)$ satisfying the comparison inequality, the exact-solution error grows at the same rate, with constant $C_G$.
Reading between the lines
- A direct numerical test is a natural next step: for $A=0$, $u\equiv1$, and linear $H(t,x)=ax$, the exact mild solution is available, so the theorem's constant $\varepsilon/(1-\tilde\lambda)$ can be compared with the actual weighted sup-distance for constructed approximate solutions.
- The same fixed-point scheme should transfer to $\psi$-Hilfer derivatives and to impulsive or neutral variants of the Cauchy problem, since the argument only needs a resolvent family and the weighted-space contraction estimate.
- The stated Lipschitz condition uses the function-space norm $\|x-y\|_{C_{1-\gamma}}$ where $x,y$ are pointwise values; a sympathetic repair would impose a pointwise Lipschitz condition with a weight $s^{\gamma-1}$ and carry the extra factor through the contraction constant, which would keep the theorems intact while making the estimates valid.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies Ulam-Hyers and Ulam-Hyers-Rassias stability for mild solutions of the fractional nonlinear abstract Cauchy problem with a Hilfer derivative, on finite intervals [0,T] and on [0,∞). The solution operator is defined via an (α,β)-resolvent family, and stability is proved by showing that the associated operator Λ is a contraction in a weighted space of continuous functions and then applying the Banach fixed point theorem. Theorems 3.1 and 3.2 treat the finite interval, Theorems 4.1 and 4.2 the half-line, and the authors note limiting cases β→0, β→1 and α=1 corresponding to Riemann-Liouville, Caputo and integer-order settings.
Significance. If the proofs were correct, the paper would provide a fairly general Ulam-Hyers stability result for Hilfer-type fractional evolution equations, with explicit stability constants and with several classical fractional-derivative settings as limiting cases. The proof strategy is standard and the organization around the resolvent operator is reasonable. The paper does not provide machine-checked proofs or reproducible code, but the explicit constants are a useful feature. However, the central contraction estimates all rely on a Lipschitz condition that is ill-typed as written, and the subsequent estimates omit weight factors that are essential for the C_{1-γ} norm; Theorem 4.2 additionally contains an undefined parameter and an internally inconsistent proof. These issues affect every main theorem, so the results as stated are not established.
major comments (3)
- [Section 2, condition (2.5)] Condition (2.5) is ill-typed: it states ||H(t,x)-H(t,y)|| ≤ ℓ(t)||x-y||_{C_{1-γ}} for x,y∈Ω, but ||·||_{C_{1-γ}} is a norm on functions on I, not on elements of Ω. If Ω is embedded as constant functions, then on I=[0,∞) the norm of a nonzero constant difference is infinite when γ<1, making the condition vacuous, and on [0,T] it introduces an extra factor T^{1-γ} that is not carried through the proofs. In the proofs of Theorems 3.1, 3.2, 4.1 and 4.2, the displayed estimate ||H(s,φ(s))-H(s,ξ(s))|| ≤ ℓ(s)||φ-ξ||_{C_{1-γ}} is used; this does not follow from (2.5) as written. If the intended hypothesis is the usual Ω-norm Lipschitz condition, the valid pointwise estimate is ||H(s,φ(s))-H(s,ξ(s))|| ≤ ℓ(s)s^{γ-1}||φ-ξ||_{C_{1-γ}}, and the missing factor s^{γ-1} appears inside the integrals in every contraction estimate. This changes the constants in (3.2), Theorem 3.2, (4.2) and Theorem 4.2, and for γ<1 the integral may even diverge at s=0 unless additional hypotheses on ℓ and u are imposed. The contraction step, on which all stability conclusions rest, is therefore not justified as written.
- [Theorem 3.2 proof] Independently of condition (2.5), the proof of Theorem 3.2 contains an unjustified replacement of the global norm ||h-g||_{C_{1-γ}} by C(h,g)φ(s). From ||t^{1-γ}(h(t)-g(t))|| ≤ C(h,g)φ(t) one may conclude pointwise that ||h(s)-g(s)|| ≤ s^{γ-1}C(h,g)φ(s), but not that the global supremum is bounded by the pointwise value C(h,g)φ(s). The displayed chain of inequalities leading to d_{φ,1-γ}(Λh,Λg) ≤ δρKT^{1-γ}d_{φ,1-γ}(h,g) therefore skips a necessary weight factor s^{γ-1} as well as a supremum argument, so the contraction in the Bielecki-type metric is not established.
- [Theorem 4.2] Theorem 4.2 is stated on the half-line [0,∞), but its proof and hypothesis (4.10) use a parameter T that is never defined in the theorem; the proof even displays t∈[0,T] at one point. Moreover, after the estimate ≤ ρC(h,g)Kφ(t), the next display introduces a factor T^{1-γ} without any justification, and the concluding bound (4.12) omits that factor. The theorem as stated is therefore not proved: the contraction constant, the fixed-point space, and the final stability bound are inconsistent.
minor comments (6)
- [Section 2] After condition (2.5), the functions ℓ and u are declared on [0,T], although later the interval I may be [0,∞); the domains should be I consistently.
- [Theorem 3.1 statement] The theorem says the resolvent operator acts on a Banach space (Ω, ||·||_{C_{1-γ}}), which conflates the state space Ω with the function space C_{1-γ}(I,Ω); S_{α,β}(t) maps Ω to Ω, while the contraction argument is in C_{1-γ}(I,Ω).
- [Theorem 3.1 proof] The sentence 'since λ~<1, Λ is a contradiction' should read 'Λ is a contraction'.
- [Theorem 3.2 proof] The completeness of (C_{1-γ}(I,Ω), d_{φ,1-γ}) is asserted but not proved; while standard, it should be justified or cited.
- [Sections 3 and 4] Several displays refer to 'Eq.(6)' (e.g., Theorem 3.4 and the discussion in Remark 3.3) although the equation numbers are not assigned to (2.9); renumber the references to the actual displayed equations.
- [Theorem 4.2 statement] The constants in the theorem and proof are written inconsistently: hypothesis (4.10) uses ρKT^{1-γ}<1, the proof derives δ_{φ,1-γ} ≤ 1/(1-T^{1-γ}ρK), and conclusion (4.12) gives 1/(1-Kρ) φ(t) without T^{1-γ}; these should be harmonized.
Circularity Check
No significant circularity: the stability theorems follow from the stated Lipschitz and contraction hypotheses by the Banach fixed point theorem, with self-citations confined to background definitions and paired with independent references.
full rationale
The derivation chain is self-contained. Theorem 3.1 assumes the contraction condition \tilde{\lambda} = \delta T^{1-\gamma}\int_0^T e^{w(T-s)}|u(s)|\ell(s)\,ds < 1 and, using the Lipschitz hypothesis (2.5) and the resolvent bounds, proves d_{1-\gamma}(\Lambda\varphi,\Lambda\xi) \le \tilde{\lambda}\,d_{1-\gamma}(\varphi,\xi), then obtains d_{1-\gamma}(\theta,v) \le \varepsilon/(1-\tilde{\lambda}) by the triangle inequality. Theorems 3.2, 4.1 and 4.2 follow the same pattern with the corresponding weighted metrics d_{\varphi,1-\gamma} and with the constants K, \rho and C_\varphi determined by the hypotheses. No parameter is fitted to data and no output quantity is an input by construction: the Ulam-Hyers constants are explicit functions of the assumed Lipschitz function, the resolvent bounds and the chosen \varphi, and the fixed point v is produced by Banach's fixed point theorem. The self-citations [1,2,5,6,8] supply background definitions, the weighted function space and auxiliary representation results; Lemma 2.1 is cited jointly with the independent reference [7], and the stability definitions are adapted from the external methodology [10]. The ill-typed Lipschitz condition (2.5) and the missing C_{1-\gamma} weight factors are genuine correctness concerns, but they are not circularity: the argument does not assume the stability conclusion it proves.
Assumptions & free parameters
assumptions (6)
- standard math Banach fixed point theorem.
- domain assumption A generates a C0-semigroup (S(t))_{t≥0}; the (α,β)-resolvent S_{α,β} and T_α satisfy the growth bounds ||S_{α,β}(t)|| ≤ δe^{wt} and ||T_α(t)|| ≤ δe^{wt}.
- domain assumption The function H satisfies the Lipschitz-type condition (2.5) with respect to the norm ||·||_{C_{1-γ}}.
- domain assumption The functions ℓ, u, and ℓu are locally integrable on I; in the infinite interval case, the relevant integrals converge.
- domain assumption In Theorems 3.2 and 4.2, there exists a continuous φ satisfying ∫_0^t φ(s)ds ≤ Kφ(t), and G is equivalent to φ in the sense αφ ≤ G ≤ βφ.
- domain assumption Lemma 2.1: Eq (1.1) is equivalent to the integral equation (2.6) and the mild solution is given by (2.7).
Cite this review
Pith. "Pith review of Ulam-Hyers stabilities of mild solutions of the fractional nonlinear abstract Cauchy problem." pith.science (2026). https://pith.science/paper/6MYCEU4F
@misc{pith2026190805296,
author = {Pith},
title = {Pith review of: Ulam-Hyers stabilities of mild solutions of the fractional nonlinear abstract Cauchy problem},
year = {2026},
howpublished = {\url{https://pith.science/paper/6MYCEU4F}},
note = {Machine review of arXiv:1908.05296}
}
abstract
Since the main work on Ulam-Hyers dependable stabilities of differential equations to date, numerous significant and applicable papers have been published, both in the sense of integer order and fractional order differential equations. However, when we enter the field of fractional differential equations, the path that is still long to be traveled, although there is a range of published works. In this sense, in this paper, we will investigate the Ulam--Hyers and Ulam--Hyers--Rassias stabilities of mild solutions of the fractional nonlinear abstract Cauchy problem on the intervals $[0,T]$ and $[0,\infty)$, by means of Banach fixed point theorem.
Reference graph
Works this paper leans on
-
[1]
Vanterler da C. Sousa, J., Kucche, Kishor D., Capelas de Oliveira, E ., Stability of ψ-Hilfer impulsive fractional differential equations, Appl. Math. Lett., 88 (2 019) 73–80
-
[2]
Sousa, J., Capelas de Oliveira, E., On the ψ-Hilfer fractional derivative, Commun
Vanterler da C. Sousa, J., Capelas de Oliveira, E., On the ψ-Hilfer fractional derivative, Commun. Nonlinear Sci. Numer. Simul., 60 (2018) 72–91
work page 2018
-
[3]
Sousa, J., Capelas de Oliveira, E., Leibniz type rule: ψ-Hilfer fractional operator, Commun
Vanterler da C. Sousa, J., Capelas de Oliveira, E., Leibniz type rule: ψ-Hilfer fractional operator, Commun. Nonlinear Sci. Numer. Simul., 77 (2019) 305–311
work page 2019
-
[4]
Chen, C., Li, M., On fractional resolvent operator functions, Se migroup Forum, 80(1) (2010) 121–142
work page 2010
-
[5]
Sousa, J., Capelas de Oliveira, E., On the Ψ-fractio nal integral and appli- cations, Comp
Vanterler da C. Sousa, J., Capelas de Oliveira, E., On the Ψ-fractio nal integral and appli- cations, Comp. Appl. Math., 38(1) (2019) 4
work page 2019
-
[6]
Existence of mild solutions to Hilfer fractional evolution equations in Banach space
Vanterler da C. Sousa, J., Existence of mild solutions to Hilfer frac tional evolution equa- tions in Banach space, arXiv:1812.02213, (2018)
work page Pith review arXiv 2018
-
[7]
J., Existence of mild solution for evolution equatio n with Hilfer frac- tional derivative, Appl
Gu, H., Trujillo, J. J., Existence of mild solution for evolution equatio n with Hilfer frac- tional derivative, Appl. Math. Comput., 257 (2015) 344–354
work page 2015
-
[8]
Vanterler da C. Sousa, J., Capelas de Oliveira, E., Ulam–Hyers stab ility of a nonlinear fractional Volterra integro-differential equation, Appl. Math. Le tt., 81 (2018) 50–56. 14
work page 2018
Show all 34 references
-
[9]
Wang, J., Zhou, Y., Mittag-Leffler-Ulam stabilities of fractional ev olution equations, Appl. Math. Lett., 25(4) (2012) 723–728
2012
-
[10]
Akkouchi, M., Bounabat, A., Rhali, M. H. L., Fixed point approach t o the stability of integral equation in the sense of Ulam-Hyers-Rassias, Annales Mat hematicae Silesianae, 24 (2011) 27–44
2011
-
[11]
Nonlinear Sci
Wang, J., Lv, L., Zhou, Y., New concepts and results in stability of fractional differential equations, Commun. Nonlinear Sci. Numer. Simul., 17(6) (2012) 2530 –2538
2012
-
[12]
Nonlinear Sci
Wang, J., Zhang, Y., A class of nonlinear differential equations wit h fractional integrable impulses, Commun. Nonlinear Sci. Numer. Simul., 19(9) (2014) 3001–3 010
2014
-
[13]
Wang, J., Zhou, Y., Fec, M., Nonlinear impulsive problems for fract ional differential equa- tions and Ulam stability, Computers & Mathematics with Applications, 6 4(10) (2012) 3389–3405
2012
-
[14]
Kilbas, A. A. A., Srivastava, H. M., Trujillo, J. J., Theory and applic ations of fractional differential equations, 204, 2006
2006
-
[15]
G., Kilbas, A
Samko, S. G., Kilbas, A. A. and Marichev, O. I., Fractional integr als and derivatives, 1993, Gordon and Breach Science Publishers, Yverdon Yverdon-les-Bain s, Switzerland
1993
-
[16]
Zhou, Y., Jiao, F., Existence of mild solutions for fractional neut ral evolution equations, Comput. Math. Appl., 59(3) (2010) 1063–1077
2010
-
[17]
Value Probl., 2019(1) (2019) 19
Saadati, R., Pourhadi, E., Samet, B., On the PC -mild solutions of abstract fractional evolution equations with non-instantaneous impulses via the measur e of noncompactness, Bound. Value Probl., 2019(1) (2019) 19
2019
-
[18]
Yang, M., Wang, Q., Existence of mild solutions for a class of Hilfer f ractional evolution equations with nonlocal conditions, Frac. Calc. Appl. Anal., 20(3) (2 017) 679–705
-
[19]
Sousa, J., Oliveira, D
Vanterler da C. Sousa, J., Oliveira, D. S., Capelas de Oliveira, E., A n ote on the mild solutions of Hilfer impulsive fractional differential equations, arXiv:1 811.09256, (20108)
-
[20]
Dabas, J., Chauhan, A., Kumar, M., Existence of the mild solutions for impulsive fractional equations with infinite delay, Inter. J. Diff. Equa., 2011 (2011)
2011
-
[21]
Jawahdou, A., Mild solutions of functional semilinear evolution Volt erra integrodifferential equations on an unbounded interval, Nonlinear Analysis: Theory, Me thods & Applications, 74(18) (2011) 7325–7332
2011
-
[22]
Balachandran, K., Annapoorani, N., Existence results for impuls ive neutral evolution in- tegrodifferential equations with infinite delay, Nonlinear Analysis: Hy brid Systems, 3(4) (2009) 674–684
2009
-
[23]
Olszowy, L., Existence of mild solutions for the semilinear nonlocal problem in Banach spaces, Nonlinear Analysis: Theory, Methods & Applications, 81 (20 13) 211–223
-
[24]
Chen, P., Li, Y., Existence of mild solutions for fractional evolutio n equations with mixed monotone nonlocal conditions, Zeitschrift f¨ ur angewandte Math ematik und Physik, 65(4) (2014) 711–728. 15
2014
-
[25]
I., Existence and Uniqueness of Mittag-Leffler-Ulam S table Solution for Frac- tional Integrodifferential Equations with Nonlocal Initial Condition s, European J
Abbas, M. I., Existence and Uniqueness of Mittag-Leffler-Ulam S table Solution for Frac- tional Integrodifferential Equations with Nonlocal Initial Condition s, European J. Pure and Appl. Math., 8(4) (2015) 478–498
2015
-
[26]
H., On the stability of the linear functional equation, P roceedings of the National Academy of Sciences of the United States of America, 2(4) (1941) 222
Hyers, D. H., On the stability of the linear functional equation, P roceedings of the National Academy of Sciences of the United States of America, 2(4) (1941) 222
1941
-
[27]
M., Problems in modern mathematics, science editions, 19 64, John-Wiley & Sons Inc
Ulam, S. M., Problems in modern mathematics, science editions, 19 64, John-Wiley & Sons Inc. New York
-
[28]
M., A collection of mathematical problems, 8 (1960)
Ulam, S. M., A collection of mathematical problems, 8 (1960)
1960
-
[29]
Aoki, T., On the stability of the linear transformation in Banach sp aces, J. Math. Soc. Jpn, 2(1-2) (1950) 64–66
1950
-
[30]
M., Homomorphisms and derivations in prope r JCQ-triples, J
Park, C., Rassias, T. M., Homomorphisms and derivations in prope r JCQ-triples, J. Math. Anal. and Appl., 337(2) (2008) 1404–1414
2008
-
[31]
M., On the stability of the linear mapping in Banach spac es, Proc
Rassias, T. M., On the stability of the linear mapping in Banach spac es, Proc. Am. Math. Soc., 72(2) (1978) 297–300
1978
-
[32]
M., Planas, G., Mild solutions to the time frac tional Navier-Stokes equations in RN , J
de Carvalho-Neto, P. M., Planas, G., Mild solutions to the time frac tional Navier-Stokes equations in RN , J. Diff. Equ., 259(7) (2015) 2948–2980
2015
-
[33]
Jarad, F., Abdeljawad, T., Generalized fractional derivatives a nd Laplace transform, Dis- crete & Continuous Dynamical Systems-S, (2019) 1775–1786
2019
-
[34]
Sousa, J., Gala, S., Capelas de Oliveira, E., On the u niqueness of mild solutions to the time-fractional Navier-Stokes equations in LN (RN )N , arXiv:1907.06587 (2019)
Vanterler da C. Sousa, J., Gala, S., Capelas de Oliveira, E., On the u niqueness of mild solutions to the time-fractional Navier-Stokes equations in LN (RN )N , arXiv:1907.06587 (2019)
2019 arXiv
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.