Direct Algorithms for Reconstructing Small Conductivity Inclusions in Subdiffusion

Bangti Jin; Jiho Hong; Zhizhang Wu

REVIEW 3 major objections 5 minor 33 references

Direct Algorithms for Reconstructing Small Conductivity Inclusions in Subdiffusion

T0 review · 3 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 develops and analyzes the first direct algebraic algorithms for recovering the locations of small conductivity inclusions in the time-fractional subdiffusion model from boundary measurements.

desk verdict Solid one-inclusion theory and numerics, but the multi-inclusion indicator rests on an unproved equivalence — still worth a referee. read the letter →

arxiv 2505.22245 v1 pith:7LN4NBW4 submitted 2025-05-28 math.NA cs.NA

Jiho Hong , Bangti Jin , Zhizhang Wu This is my paper

classification math.NAcs.NA MSC 35R3065M32

keywords subdiffusioninverseconductivityproblemsmallinclusionsdirectreconstructionalgorithmasymptoticexpansionpolarizationtensorfractionaldiffusionboundarymeasurement

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

The paper targets an inverse conductivity problem for subdiffusion, the model in which a Caputo fractional derivative of order $\alpha\in(0,1)$ replaces the usual time derivative. It aims to locate small well-separated inclusions of anomalous conductivity from a weighted boundary measurement $I_\Phi$, without running any iterative optimization. The central result is an asymptotic expansion of $I_\Phi$ in powers of the inclusion size $\varepsilon$, whose leading term is a sum over inclusions of the polarization tensor contracted with background and test gradients. On top of this the paper builds two direct algorithms, one for a single inclusion and one for multiple inclusions, using truncated asymptotic fundamental solutions, and proves an $O(\varepsilon\,\ell_{\varepsilon,d}^{1/2})$ location-error bound for the single-inclusion case. A sympathetic reader should care because these are, by the authors' account, the first direct algorithms for this fractional inverse problem, and the numerics show accurate reconstructions for circular and elliptical inclusions despite the absence of any forward solve.

What carries the argument

The load-bearing identity is the asymptotic expansion (2.7) of Theorem 2.1, with the polarization tensor $M^{(\ell)}$ of the rescaled inclusion $B_\ell$ as the coefficient; for a disk, $M^{(\ell)}=-\frac{d\gamma_0|B_\ell|}{\gamma_\ell+(d-1)\gamma_0}I_d$. The computational engine is the approximate fundamental solution $\Psi_{(x_0,0),N}(x,t)=(\gamma_0 t^\alpha)^{-d/2}\Psi_{d,\alpha,N}(|x-x_0|/\sqrt{\gamma_0 t^\alpha})$, the truncated far-field expansion of the reduced Green function, whose gradient has the explicit radial form $\nabla\Psi_{d,\alpha,N}(x)=x\,S_{d,N}(|x|^2)$. The single-inclusion algorithm zeros $I_\Phi$ for two independent harmonic backgrounds and intersects the resulting projection sets; the multi-inclusion algorithm forms the response matrix $B_I$ and compares $\|G_n(z)\|_F$ with $\|Q_{n,k}G_n(z)\|_F$. The proofs rest on fractional integration by parts, a coercivity inequality for the Caputo derivative, a fractional-energy stability bound for the perturbation $V$, and singular-value perturbation estimates for the compact operator $M$.

What would settle it

Apply the multi-inclusion indicator to data from a homogeneous domain containing no inclusions; if the indicator develops a local maximum or grows large at any interior point, then the surface-ratio criterion is wrong and the Section 4 reconstructions have no theoretical guarantee.

Watch

Extended reading notes

Core claim

The central claim is Theorem 2.1: for $m$ small, well-separated inclusions $A_\ell=\varepsilon B_\ell+z_\ell$, the boundary measurement $I_\Phi=\int_0^T\int_{\partial\Omega}\gamma_0(u-U)\partial_n\Phi\,d\sigma\,dt$ satisfies $$I_\Phi = -\varepsilon^d\sum_{\ell=1}^m(\gamma_0-\gamma_\ell)\int_0^T \nabla U(z_\ell,t)\cdot $M^{{(\ell)}}$\nabla\Phi(z_\ell,t)\,dt + O(\ell_{\varepsilon,d}^{1/2}\$varepsilon^{{d+1}}$),$$ provided the test function $\Phi$ solves the adjoint fractional problem with terminal value zero. The paper then shows that using a truncated far-field expansion of the fundamental solution, rather than the exact solution, perturbs $I_\Phi$ by only $O(\varepsilon^{d/2})$ when the source is far enough and the truncation level $N$ is large enough, so the asymptotic law remains valid. That yields Theorem 3.6: the single-inclusion algorithm recovers the center with $|P-z_1|\le C\varepsilon\,\ell_{\varepsilon,d}^{1/2}$. For multiple inclusions, the paper constructs a matrix of pairwise responses and a discrete indicator $W_{n,k}(z)$ that approximates the idealized ratio $\varphi(z)=\|L_z\|_{\mathrm{HS}}/\|(\mathrm{id}-P_M)L_z\|_{\mathrm{HS}}$, whose blow-up is used to mark inclusion centers. The authors claim these are the first direct algorithms for the inverse conductivity problem in the fractional subdiffusion model.

Load-bearing premise

The multi-inclusion algorithm depends on an unproved equivalence: an interior point is called an inclusion center exactly when a certain surface-measurement ratio becomes infinite, whereas the paper actually proves a related condition on integrals over a small ball around the point.

Editorial extensions

If this is right

For one inclusion, the reconstruction error is provably $O(\varepsilon\,\ell_{\varepsilon,d}^{1/2})$, so smaller inclusions are located with proportionally better absolute accuracy, although the relative signal-to-noise worsens.
The algorithms avoid all forward solves and optimization: the single-inclusion case reduces to two scalar root-finding steps and an intersection of lines or cylinders, and the multi-inclusion case reduces to forming a response matrix and comparing Frobenius norms of two matrices.
The truncation level $N$ and source distance $x_0$ must satisfy Assumption 3.1, meaning the approximate fundamental solution is accurate enough that its perturbation stays at order $\varepsilon^{d/2}$; with this choice the leading-order measurement law (2.7) survives unchanged.
Although the theory is stated for circular inclusions, the numerical experiments show the same algorithms locate elliptical inclusions, consistent with the fact that the asymptotic expansion only uses the inclusion's polarization tensor.
The integral measurement $I_\Phi$ smooths boundary noise, so the reconstructions remain stable for a few percent noise even though the leading-order signal is only of order $\varepsilon^d$.

Reading between the lines

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

The same algebraic pipeline should transfer to partial-aperture or later-time-window measurements, because the source-distance and truncation conditions are local and the adjoint terminal condition can be re-solved with the same truncated fundamental solution.
Because the leading-order term is linear in the combined quantities $(\gamma_0-\gamma_\ell)M^{(\ell)}$, a single boundary data set cannot separately determine conductivity contrast, size, and shape; extracting the individual parameters would require multiple experiments or extra assumptions.
A natural stress test not performed in the paper is to run the multi-inclusion indicator on a homogeneous sample with no inclusions: if any interior peak appears, the surface-degeneracy criterion would need modification.

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.

Desk Editor's Note

read the letter

This paper does something real: it extends the Ammari–Iakovleva–Kang–Kim direct algorithms from the heat equation to the Caputo subdiffusion model, and adds 3D constructions. The central asymptotic expansion, Theorem 2.1, is derived carefully using Alikhanov's inequality, and the one-inclusion error bound in Theorem 3.6 looks right. The analysis of the truncated fundamental solution (the interplay between N, dist(x0, Ω), and ε) is a genuine technical contribution that should transfer to other fractional inverse problems. Numerics for one inclusion are convincing, including elliptical inclusions and noise. The soft spot is exactly where the stress-test note lands. The multi-inclusion indicator φ(z) = ||Lz||_HS / ||(id − PM)Lz||_HS is asserted to characterize inclusion centers, but that equivalence is never proved. Theorem 3.7 gives a domain-integral condition — if a cancellation limit holds for every f, then z is in Z — and the appendix proof of that theorem is fine. But the jump from that condition to the surface-degeneracy ratio is a separate mathematical claim, and no theorem states or proves it. Theorem 3.10 only shows that the discrete W_{n,k} approximates φ_k(z) as n → ∞; it inherits the gap rather than filling it. So the multi-inclusion reconstructions in Section 4 are currently a heuristic with nice numerics, despite the paper's claim of "theoretical underpinnings" for the algorithm. That should be fixed or explicitly softened before the paper's full claims are accepted. Minor: Lemma 2.3's proof is omitted, though the result is standard and the omission is not damaging. No code is released, which is a shame for a numerical algorithms paper. The citation pattern looks honest; the relevant heat-equation and fractional literature is cited. Who is this for? Researchers in fractional inverse problems and anyone doing direct imaging in anomalous transport. It deserves a serious referee. If I were the editor, I would send it out and ask the authors to either prove the multi-inclusion indicator equivalence or state plainly that it is an empirical heuristic.

Referee Report

3 major / 5 minor

Summary. The paper develops direct, algebraically cheap reconstruction algorithms for locating small conductivity inclusions in a time-fractional (Caputo) subdiffusion model from boundary measurements. The central theoretical result is an asymptotic expansion (Theorem 2.1) of the weighted boundary measurement I_Φ in terms of the inclusion size ε, polarization tensors, and gradients of the background and test solutions. Based on this expansion, the authors propose a one-inclusion algorithm using harmonic background solutions and truncated approximate fundamental solutions, with an error estimate (Theorem 3.6), and a multi-inclusion algorithm based on a matrix indicator function related to a factorization-method-type operator (Theorems 3.7, 3.8, 3.10). Numerical experiments in two dimensions illustrate the performance for circular and elliptical inclusions under noise.

Significance. If the results hold, this is the first direct (non-iterative) reconstruction method for small inclusions in the subdiffusion model, extending the heat-equation algorithms of Ammari et al. [3] to the fractional setting. The asymptotic expansion Theorem 2.1 is derived in detail against independent mathematical inputs (polarization tensors, fundamental solution asymptotics), with no parameters fitted to the target reconstructions; the one-inclusion error bound in Theorem 3.6 is a genuine quantitative guarantee. The main weakness is that the multi-inclusion indicator, which is central to Section 3.3 and the corresponding numerical section, is not rigorously connected to the proven characterization in Theorem 3.7.

major comments (3)

[Section 3.3, after Theorem 3.7] The paper defines the indicator φ(z) = ||L_z||_HS / ||(id − P_M)L_z||_HS and states that 'Theorem 3.7 suggests that an inclusion is located at z if φ(z) blows up,' but no theorem proves this equivalence. Theorem 3.7 establishes a domain-integral criterion (3.19) in terms of integrals over B_ρ(z)\A; for z ∈ Z the criterion holds trivially because B_ρ(z)\A is empty for small ρ, and the Appendix proof shows only that non-membership gives a positive limit for a carefully chosen f. Nothing in the paper shows that the surface Hilbert-Schmidt ratio φ(z) has peaks exactly at the inclusion centers. Theorem 3.10 merely shows that the discrete indicator W_{n,k}(z) approximates φ_k(z) as n → ∞, so it inherits, rather than fills, this gap. Consequently, the multi-inclusion reconstructions in Section 4.2 and the claim of theoretical underpinnings for the multiple-inclusion algorithm are not supported by the presented analysis.
[Lemma 2.3] The proof of Lemma 2.3, which gives the integral representation I_Φ = Σ (γ0−γℓ) ∫_{A_ℓ} ∇u·∇Φ dx dt, is omitted with the remark that it is similar to [3, Lemma 3.3]. Since this representation is an essential ingredient in the proof of the main asymptotic expansion Theorem 2.1, the proof should be included in full or a precise reference with all assumptions verified should be supplied.
[Theorem 3.10, proof] The proof invokes Lemma 3.9 to control the difference between the singular projectors of M and M_n, but the hypothesis of Lemma 3.9 is lim_{n→∞} ||M − M_n|| = 0. In the proof of Theorem 3.10, the authors only establish lim_{n→∞} ||M − M_n|| = O(ε^{d+1}ℓ^{1/2}_{ε,d}), which does not tend to zero for fixed ε. A quantitative version of Lemma 3.9 with an explicit error term, or a different perturbation argument, is needed to justify the bound (3.26). As written, the approximation of φ_k by the discrete ratio W_{n,k} is not fully established.

minor comments (5)

[Section 1] There is a typo in the second sentence: 'Supose' should be 'Suppose'.
[Theorem 3.4, Eq. (3.10)] The inequalities in (3.10) are written with the sign of r as if r ≥ 0. Since r = −(γ0−γ1)∇U(z1)·(z1−x0)/|z1−x0| can be negative depending on the directions, the statement should either assume r > 0 or use |r| in the bounds and state the corresponding sign properties separately.
[Section 3.3, after Theorem 3.7] In the definition of the idealized indicator, the set A in B_ρ(z)\A depends on ε; this dependence should be made explicit in the notation, for example by writing A_ε.
[Section 4] The numerical section states that a Galerkin FEM with linear elements and the L1 scheme are used, with 2^7 uniform time subintervals, but no spatial mesh size or adaptivity parameters are reported; adding these details would improve reproducibility.
[Lemma 3.3] The condition (3.3) N ≥ ⌈δ|log ε|⌉ involves the truncation level N, but in the numerical experiments N is fixed at 3. A brief remark on the range of ε for which this choice is consistent with Assumption 3.1 would be helpful.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the asymptotic expansion and one-inclusion algorithm are derived from independent mathematical inputs, and the multi-inclusion indicator gap is a completeness issue, not a circular reduction.

full rationale

The central result, Theorem 2.1, derives the boundary measurement expansion from an integral representation (Lemma 2.3), an a priori estimate for the perturbed problem (Lemma 2.4), and polarization tensors defined by independent cell problems. No parameter is fitted to the inclusion locations or to the target expansion. Theorem 3.4 extends the expansion to approximate fundamental solutions using the external asymptotic results of Qiu and Sim [32] and a rigorous perturbation estimate (Lemma 3.3). The one-inclusion reconstruction error bound in Theorem 3.6 follows from the derived two-sided inequality (3.10), and the zero-finding equation (3.12) is motivated by the leading-order term rather than engineered to encode the answer. For multiple inclusions, Theorem 3.7 proves only a domain-integral sufficient condition; the subsequent passage to the surface-degeneracy ratio phi(z) is explicitly presented as 'Theorem 3.7 suggests' rather than as a proven equivalence, and Theorem 3.10 only shows that the discrete indicator approximates phi_k(z). This is an unproven heuristic leap (a correctness gap), not a circular reduction, because phi(z) is defined independently of the set Z of centers and the numerical experiments in Section 4 validate the algorithm against full-model synthetic data rather than against the asymptotic formula itself. Self-citations, including Jin's monograph [16] and numerical papers [9,17,18,19], are used for standard fractional-calculus facts, background uniqueness results, and forward solvers; none of the load-bearing mathematical claims reduces to an unverified self-cited theorem. The derivation chain is therefore self-contained with respect to circularity, even though the multiple-inclusion theory is incomplete.

Assumptions & free parameters 2 free parameters · 6 assumptions · 0 invented entities

No free parameters are used in the derivation of the asymptotic expansion; the only hand-chosen quantities are algorithm hyperparameters N and k. The central expansion rests on standard fractional calculus identities and cited results on fundamental solutions and polarization tensors. The paper introduces no new physical entities.

free parameters (2)

N (truncation order of approximate fundamental solution) = 3 in all numerical experiments
Algorithm parameter; theory requires N >= delta |log epsilon|, but experiments fix N=3 without verifying the assumption. Not fitted to the target, but hand-chosen.
k (rank truncation in W_{n,k}) = 5, 7, or 5-9 depending on singular values
Chosen from the singular value decay of the data matrix B_I in each experiment; this is a data-adaptive free parameter in the multi-inclusion indicator.

assumptions (6)

standard math Alikhanov's inequality w partial_t^alpha w >= (1/2) partial_t^alpha w^2
Invoked in Lemma 2.2 to derive the energy estimate; cited to [2, Lemma 1].
standard math Fractional integration-by-parts formula for Caputo derivatives
Used in Lemma 2.3 to represent I_Phi; cited to [16, Lemma 2.6].
domain assumption Solvability and regularity of the forward subdiffusion initial boundary value problem
Used throughout to justify operations on u and U; cited to [29, Theorems 3.5, 3.8, 3.11].
domain assumption Asymptotic expansion of the reduced Green function Psi_{d,alpha} with coefficients a0 and a_{j,k}
Basis for the approximate fundamental solutions in Sections 3.1-3.3; cited to [32, Theorems 3.1 and 3.6].
domain assumption Pola-Szego polarization tensor properties and the disk formula for M^(l)
Used in the leading-order term (2.7) and the sign analysis in Theorem 3.4; cited to [4, Chapter 3].
domain assumption Pointwise bounds on derivatives of the fractional fundamental solution near the source
Used in Appendix A to control singular integrals in Theorem 3.7; cited to [22, Theorems 2.4 and 2.5].

how reviews work

0 comments

Cite this review

Pith. "Pith review of Direct Algorithms for Reconstructing Small Conductivity Inclusions in Subdiffusion." pith.science (2026). https://pith.science/paper/7LN4NBW4

@misc{pith2026250522245,
  author       = {Pith},
  title        = {Pith review of: Direct Algorithms for Reconstructing Small Conductivity Inclusions in Subdiffusion},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7LN4NBW4}},
  note         = {Machine review of arXiv:2505.22245}
}

read the original abstract

The subdiffusion model that involves a Caputo fractional derivative in time is widely used to describe anomalously slow diffusion processes. In this work we aim at recovering the locations of small conductivity inclusions in the model from boundary measurement, and develop novel direct algorithms based on the asymptotic expansion of the boundary measurement with respect to the size of the inclusions and approximate fundamental solutions. These algorithms involve only algebraic manipulations and are computationally cheap. To the best of our knowledge, they are first direct algorithms for the inverse conductivity problem in the context of the subdiffusion model. Moreover, we provide relevant theoretical underpinnings for the algorithms. Also we present numerical results to illustrate their performance under various scenarios, e.g., the size of inclusions, noise level of the data, and the number of inclusions, showing that the algorithms are efficient and robust.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

33 extracted references · 31 canonical work pages

[3]

Ammari, E

H. Ammari, E. Iakovleva, H. Kang, and K. Kim , Direct algorithms for thermal imaging of small inclusions, Multiscale Model. Simul., 4 (2005), pp. 1116–1136

work page 2005
[1]

Aceto and F

L. Aceto and F. Durastante , Efficient computation of the Wright function and its applications to fractional diffusion-wave equations , ESAIM: Math. Model. Numer. Anal., 56 (2022), pp. 2181– 2196. 30

work page 2022
[2]

Alikhanov, A priori estimates for solutions of boundary value problems for fractional-order equa- tions, Differ

A. Alikhanov, A priori estimates for solutions of boundary value problems for fractional-order equa- tions, Differ. Equ., 46 (2010), pp. 660–666

work page 2010
[4]

Ammari and H

H. Ammari and H. Kang , Reconstruction of Small Inhomogeneities from Boundary Measurements , vol. 1846 of Lecture Notes in Mathematics, Springer-Verlag, Berlin, 2004

work page 2004
[5]

, Polarization and Moment Tensors , Springer, New York, 2007

work page 2007
[6]

Berkowitz, A

B. Berkowitz, A. Cortis, M. Dentz, and H. Scher , Modeling non-fickian transport in geological formations as a continuous time random walk , Rev. Geophys., 44 (2006)

work page 2006
[7]

Bouraoui, L

M. Bouraoui, L. El Asmi, and A. Khelifi , On an inverse boundary problem for the heat equation when small heat conductivity defects are present in a material , ZAMM Z. Angew. Math. Mech., 96 (2016), pp. 327–343

work page 2016
[8]

Bouraoui, L

M. Bouraoui, L. El Asmi, and A. Khelifi , Reconstruction of polygonal inclusions in a heat con- ductive body from dynamical boundary data , ESAIM Math. Model. Numer. Anal., 51 (2017), pp. 949–964

work page 2017

Show all 33 references

[9]

S. Cen, B. Jin, Y. Liu, and Z. Zhou, Recovery of multiple parameters in subdiffusion from one lateral boundary measurement, Inverse Problems, 39 (2023), pp. 104001, 31

2023
[10]

Cheng, J

J. Cheng, J. Nakagawa, M. Yamamoto, and T. Yamazaki, Uniqueness in an inverse problem for a one-dimensional fractional diffusion equation , Inverse Problems, 25 (2009), pp. 115002, 16

2009
[11]

D. K. Crane, M. S. Gockenbach, and M. J. Roberts , Approximating the singular value expansion of a compact operator , SIAM J. Numer. Anal., 58 (2020), pp. 1295–1318

2020
[12]

Dentz, A

M. Dentz, A. Cortis, H. Scher, and B. Berkowitz , Time behavior of solute transport in het- erogeneous media: transition from anomalous to normal transport , Adv. Water Res., 27 (2004), pp. 155–173

2004
[13]

S. D. Eidelman and A. N. Kochubei, Cauchy problem for fractional diffusion equations, J. Differential Equations, 199 (2004), pp. 211–255

2004
[14]

L. C. Evans and R. F. Gariepy, Measure Theory and Fine Properties of Functions, CRC Press, Boca Raton, FL, revised ed., 2015

2015
[15]

Gohberg, S

I. Gohberg, S. Goldberg, and M. A. Kaashoek , Classes of Linear Operators. Vol. I , Birkh¨ auser Verlag, Basel, 1990

1990
[16]

Jin , Fractional Differential Equations: an Approach via Fractional Derivatives , Springer, Cham, 2021

B. Jin , Fractional Differential Equations: an Approach via Fractional Derivatives , Springer, Cham, 2021

2021
[17]

B. Jin, X. Lu, Q. Quan, and Z. Zhou , Numerical recovery of the diffusion coefficient in diffusion equations from terminal measurement . Preprint, arXiv:2405.10708, 2024

2024 arXiv
[18]

Jin and Z

B. Jin and Z. Zhou , Numerical estimation of a diffusion coefficient in subdiffusion , SIAM J. Control Optim., 59 (2021), pp. 1466–1496

2021
[19]

, Numerical Treatment and Analysis of Time-Fractional Evolution Equations , Springer Nature, Cham, 2023

2023
[20]

Y. Kian, Z. Li, Y. Liu, and M. Yamamoto , The uniqueness of inverse problems for a fractional equation with a single measurement , Math. Ann., 380 (2021), pp. 1465–1495

2021
[21]

A. A. Kilbas, H. M. Srivastava, and J. J. Trujillo, Theory and Applications of Fractional Differ- ential Equations , Elsevier Science B.V., Amsterdam, 2006

2006
[22]

Kim and S

K.-H. Kim and S. Lim, Asymptotic behaviors of fundamental solution and its derivatives to fractional diffusion-wave equations, J. Korean Math. Soc., 53 (2016), pp. 929–967

2016
[23]

Kirsch and N

A. Kirsch and N. Grinberg , The Factorization Method for Inverse Problems , Oxford University Press, Oxford, 2008

2008
[24]

S. C. Kou and X. S. Xie, Generalized Langevin equation with fractional Gaussian noise: subdiffusion within a single protein molecule , Phys. Rev. Lett., 93 (2004), p. 180603

2004
[25]

Kubica, K

A. Kubica, K. Ryszewska, and M. Yamamoto , Time-Fractional Differential Equations—A Theo- retical Tntroduction, Springer, Singapore, 2020

2020
[26]

Kulpa, The Poincar´ e-Miranda theorem, Amer

W. Kulpa, The Poincar´ e-Miranda theorem, Amer. Math. Monthly, 104 (1997), pp. 545–550

1997
[27]

O. Kwon, J. K. Seo, and J.-R. Yoon , A real-time algorithm for the location search of discontinuous conductivities with one measurement , Comm. Pure Appl. Math., 55 (2002), pp. 1–29

2002
[28]

G. Li, D. Zhang, X. Jia, and M. Yamamoto , Simultaneous inversion for the space-dependent diffu- sion coefficient and the fractional order in the time-fractional diffusion equation, Inverse Problems, 29 (2013), pp. 065014, 36

2013
[29]

J. Mu, B. Ahmad, and S. Huang , Existence and regularity of solutions to time-fractional diffusion equations, Comput. Math. Appl., 73 (2017), pp. 985–996

2017
[30]

Podlubny, Fractional Differential Equations, Academic Press, Inc., San Diego, CA, 1999

I. Podlubny, Fractional Differential Equations, Academic Press, Inc., San Diego, CA, 1999

1999
[31]

P ´olya and G

G. P ´olya and G. Szeg ¨o, Isoperimetric Inequalities in Mathematical Physics , Princeton University Press, Princeton, 1951

1951
[32]

Qiu and J

L. Qiu and J. Sim, A direct sampling method for time-fractional diffusion equation , Inverse Problems, 31 40 (2024), pp. 065006, 33

2024
[33]

Ritchie, X.-Y

K. Ritchie, X.-Y. Shan, J. Kondo, K. Iwasawa, T. Fujiwara, and A. Kusumi , Detection of non- brownian diffusion in the cell membrane in single molecule tracking , Biophys. J., 88 (2005), pp. 2266–2277

2005

Pith tools

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

[1] [3]

Ammari, E

H. Ammari, E. Iakovleva, H. Kang, and K. Kim , Direct algorithms for thermal imaging of small inclusions, Multiscale Model. Simul., 4 (2005), pp. 1116–1136

work page 2005

[2] [1]

Aceto and F

L. Aceto and F. Durastante , Efficient computation of the Wright function and its applications to fractional diffusion-wave equations , ESAIM: Math. Model. Numer. Anal., 56 (2022), pp. 2181– 2196. 30

work page 2022

[3] [2]

Alikhanov, A priori estimates for solutions of boundary value problems for fractional-order equa- tions, Differ

A. Alikhanov, A priori estimates for solutions of boundary value problems for fractional-order equa- tions, Differ. Equ., 46 (2010), pp. 660–666

work page 2010

[4] [4]

Ammari and H

H. Ammari and H. Kang , Reconstruction of Small Inhomogeneities from Boundary Measurements , vol. 1846 of Lecture Notes in Mathematics, Springer-Verlag, Berlin, 2004

work page 2004

[5] [5]

, Polarization and Moment Tensors , Springer, New York, 2007

work page 2007

[6] [6]

Berkowitz, A

B. Berkowitz, A. Cortis, M. Dentz, and H. Scher , Modeling non-fickian transport in geological formations as a continuous time random walk , Rev. Geophys., 44 (2006)

work page 2006

[7] [7]

Bouraoui, L

M. Bouraoui, L. El Asmi, and A. Khelifi , On an inverse boundary problem for the heat equation when small heat conductivity defects are present in a material , ZAMM Z. Angew. Math. Mech., 96 (2016), pp. 327–343

work page 2016

[8] [8]

Bouraoui, L

M. Bouraoui, L. El Asmi, and A. Khelifi , Reconstruction of polygonal inclusions in a heat con- ductive body from dynamical boundary data , ESAIM Math. Model. Numer. Anal., 51 (2017), pp. 949–964

work page 2017