Pith. sign in

REVIEW 2 major objections 3 minor 37 references

Parallelisation of Discrete Exterior Calculus via Representation Theory on Curved and Three-Dimensional Meshes

T0 review · 2 major / 3 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read On any mesh with a finite symmetry group, the two fundamental operators of Discrete Exterior Calculus—the exterior derivative and the Hodge star—are equivariant, so one precomputed symmetry-adapted basis block-diagonalizes every operator bu

desk verdict The main theorem overclaims: Theorem 4.3 misses an orientation sign for orientation-reversing isometries, so universal block-diagonalization does not follow for ⋆-based operators; the Laplacian results likely survive. read the letter →

arxiv 2607.25192 v1 pith:CSTJPUYA submitted 2026-07-28 math.NA cs.NAmath.DGphysics.comp-ph

Leon D. da Silva , Marcelo P. Santos , José D. da Silva , Gilson Ferreira Jr This is my paper
classification math.NAcs.NAmath.DGphysics.comp-ph MSC 65N3065N2265Y0565Y2068W1020C1520C3558A14
keywords DiscreteExteriorCalculusequivarianceblock-diagonalizationsymmetry-adaptedbasisHodgestarrepresentationtheoryisometricgroupactionsparallelsolvers
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 proves that the discrete exterior derivative and the discrete Hodge star, the two building blocks of Discrete Exterior Calculus (DEC), commute with any finite group of isometries acting on a simplicial mesh. Because every DEC operator—the Hodge Laplacian, the codifferential, Maxwell-type operators, elasticity operators, and so on—is assembled from these two pieces, a single symmetry-adapted basis, computed once per mesh, simultaneously block-diagonalizes all of them into independent subproblems, one per irreducible representation of the symmetry group. The result extends beyond flat Platonic domains to curved surfaces like geodesic spheres and tori, and to three-dimensional tessellations, where the main obstacle is not equivariance but the invertibility of the Hodge star. The paper verifies machine-precision block-diagonalization on an icosahedral sphere, a hexagonal torus, and a body-centred-cubic tessellation of the 3-torus, reporting FLOP-based speedups up to 62x and 182x over dense factorization. If correct, this turns a single mesh-symmetry analysis into a universal, embarrassingly parallel solver strategy for structure-preserving discretizations.

What carries the argument

The engine is the permutation representation ρ_k induced by the group action on each cochain space of discrete k-forms, together with the character-theoretic projectors Π_i = (n_i/|G|) Σ_g χ_i(g⁻¹) ρ_k(g) that extract orthonormal symmetry-adapted bases. A standard intertwinement lemma then forces every equivariant operator to block-diagonalize in such a basis, with blocks indexed by irreps; because both d and ⋆ are equivariant, the same basis simultaneously works for all DEC operators on the mesh. The construction is purely combinatorial and dimension-agnostic, requiring only the combinatorial indexing of the group action and no polynomial spatial algebra.

What would settle it

Compute the discrete screened Poisson eigenvalues on the fan-BCC 3-torus using the centroid-based star at increasing resolution and compare them to the known spectrum of the flat 3-torus Laplacian; failure to converge would show that the 3D block-diagonalization is not solving the intended PDE. In 2D, a cocircular (degenerate Delaunay) mesh such as a quad grid split by a single diagonal produces a zero cotangent weight and a singular Hodge star, which would break the invertibility hypothesis and the framework.

Watch

Extended reading notes

Core claim

The paper's central claim is that under a finite group G acting by isometries on a simplicial complex, every linear operator assembled from the discrete exterior derivative d, the Hodge star ⋆, and their inverses is G-equivariant, provided each participating Hodge star is invertible. Consequently, in a symmetry-adapted basis built once from the group action—via character-theoretic projection onto isotypic components—the matrix of any such operator is exactly block-diagonal, with one block per irreducible representation. The proof establishes equivariance of d by pure combinatorics (the group action commutes with the boundary operator) and equivariance of ⋆ by the isometry condition (volume r

Load-bearing premise

The whole theorem relies on each Hodge star being invertible and on the dual complex transforming equivariantly under the symmetry; on surfaces this requires strictly Delaunay meshes to avoid vanishing cotangent weights, and in 3D the paper must switch to a centroid-based star that is positive and equivariant but is explicitly not shown to be a consistent discretization—if that star does not approximate the continuum operator, the 3D experiments show structure, not a usable s

Editorial extensions

If this is right

  • One mesh-symmetry analysis simultaneously block-diagonalizes the scalar Laplacian, the vector Laplacian, Maxwell-type operators, elasticity operators, and any other operator assembled from d and ⋆.
  • The equivariance of the exterior derivative is purely topological and dimension-agnostic, so the framework extends to three-dimensional complexes without new algebraic machinery; the three-dimensional bottleneck is Hodge-star invertibility, not equivariance.
  • Block solves are independent, enabling embarrassingly parallel factorization; measured FLOP-based parallel speedups reach 62× on the icosahedral sphere and 182× on the hexagonal torus, approaching the regular-representation asymptote.
  • Eigenfunctions of DEC operators carry definite irreducible-representation labels, giving a free symmetry classification of the discrete spectrum (verified against spherical-harmonic subduction predictions on the sphere).
  • The block-diagonalization is a property of the operator, not of the data, so it persists for anisotropic right-hand sides, inhomogeneous boundary conditions, and the implicit linear substeps of operator-splitting schemes for nonlinear PDEs.

Reading between the lines

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

  • If the centroid-based Hodge star used on the 3-torus is not a consistent discretization, then the 3D numerical section demonstrates algebraic block structure but not a convergent solver; a convergence test against the known continuum spectrum of the flat 3-torus would settle this.
  • The orbit-averaging technique for restoring equivariance on approximately symmetric meshes suggests a practical route to near-symmetry problems, but the paper gives no perturbation bounds on how block-coupling decays with symmetry error; a quantitative bound would widen applicability.
  • Since each isotypic block factorizes as a tensor product M ⊗ I, a solver that stores only one copy per irreducible representation could reduce memory and flops by a further factor close to the group order, beyond the already-reported asymptotes.
  • The machine-precision match between measured eigen-degrees and subduction predictions suggests the same basis could be used as a mesh-quality or symmetry-fidelity diagnostic on curved meshes.
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.

Referee Report

2 major / 3 minor

Summary. The paper develops a representation-theoretic block-diagonalization framework for Discrete Exterior Calculus (DEC) operators on simplicial meshes with finite isometry groups. The main theoretical claims are: (i) the discrete exterior derivative d is equivariant under simplicial automorphisms (Theorem 4.1); (ii) the discrete Hodge star ⋆ is equivariant under isometric actions (Theorem 4.3); and (iii) any operator assembled from d, ⋆, and ⋆^{-1} is block-diagonal in a single symmetry-adapted basis (Theorem 5.1). Numerical experiments on an icosahedral sphere (I_h), a hexagonal torus (D_6h), and a BCC tessellation of the flat 3-torus (T_d) report machine-precision equivariance residuals and FLOP-based speedups up to 182× (FLOP predicted) and about 100× (wall-clock).

Significance. If the equivariance results hold, the framework would be a valuable, PDE-agnostic tool for symmetry exploitation in DEC on curved and three-dimensional meshes, complementing spectral and FEEC approaches. The paper is clearly written, includes reproducible code, and the spherical-harmonic subduction tests (Table 2) provide a strong external falsifiable benchmark. The main weakness is that the Hodge-star equivariance theorem omits an orientation sign, and the numerical residuals reported for ⋆ are only consistent with a non-standard action on dual cochains. This is a load-bearing gap in the central claim, though it appears fixable by restricting the statement or by explicitly adopting and justifying a modified dual representation.

major comments (2)
  1. [§4.2, Theorem 4.3; also Theorem 5.1] The proof of Theorem 4.3(i) asserts g(⋆σ) = ⋆(gσ) as an equality of oriented cells. Under the standard DEC convention used in the paper, namely [σ, ⋆σ] = +1 (Definition 3.8 and Remark 3.7), an orientation-reversing isometry g satisfies g(⋆σ) = −⋆(gσ) because the determinant of g enters the induced orientation of the image cell. Consequently, the second equality in Eq. (27) misses a factor det(g) = −1 for reflections/inversions, and the correct relation is ⋆^k ρ^k(g) = det(g) ρ^{n−k}(g) ⋆^k (equivalently, ρ^{n−k}(g) ⋆^k = det(g) ⋆^k ρ^k(g)). Thus Theorem 4.3(ii) is false as stated. This also invalidates Theorem 5.1(i) for the star itself and for any operator assembled with an odd number of star factors. Operators such as δ and Δ, which contain two star factors, are still equivariant because the two determinant signs cancel, but the claimed 'universal block-diagonalization of every operato
  2. [§6.1, Table 1; §6.5, Table 5] The reported machine-precision equivariance residuals for ⋆ (Table 1 shows 5.9×10^{-15} for I_h, which contains reflections; Table 5 similarly for T_d) are not compatible with the standard geometric pullback on dual cochains under the convention [σ, ⋆σ] = +1. For orientation-reversing elements the standard pullback would produce a commutator defect of size ~1, not ~10^{-15}. The implementation must therefore be using a different action on dual cochains, e.g., ρ^{n−k}(g)β evaluated at ⋆σ as β(⋆(g^{-1}σ)) rather than β(g^{-1}(⋆σ)). The paper never defines the representation on Ω^{n−k}(⋆K); Theorem 4.3 assumes it. The authors need to (a) state explicitly the action on dual cochains, (b) prove that it is a representation and that it is compatible with the DEC inner product and the adjointness formula (16), and (c) justify why this action, rather than the standard pullback, is the correct one
minor comments (3)
  1. [§5.1, Step 5] Typo: 'assurres' should be 'assures'.
  2. [§6.4, Figure 6] The measured wall-clock speedup for the torus plateaus near 100× at N_0 = 3888 while the FLOP prediction continues to rise to 182×. The comment in the text is honest, but the cause of the plateau (memory bandwidth, BLAS overhead, block density?) would be useful to diagnose, since the gap is large at the largest resolution.
  3. [§6.5] The paper correctly discloses that the centroid-based Hodge star on the non-well-centered fan-BCC mesh is not shown to be a consistent discretization. That limitation is acceptable for the structural claim, but it should be restated in the abstract or conclusion where the 3D speedup is advertised, so readers do not infer a validated 3D solver.

Circularity Check

0 steps flagged · score 1.0 of 10

No circular derivation; central claims are proved from operator definitions and verified against external benchmarks.

full rationale

The paper's derivation chain is self-contained. Theorem 4.1 derives d-equivariance from the pullback definition (13) and the fact that simplicial isomorphisms commute with the boundary operator (20). Theorem 4.3 derives Hodge-star equivariance from the isometry hypothesis and the dual assignment property g(⋆σ)=⋆(gσ), with the orbit-preserved volume ratio (26); no fitted parameter enters. Theorem 5.1 then applies Schur's lemma to conclude block-diagonalization, which is standard representation theory rather than a restatement of the paper's inputs. The numerical experiments test rather than define the theory: the equivariance residuals are machine-precision checks, and Table 2's subduction predictions are external spherical-harmonic benchmarks not used to set any constant. The only self-citation, PySymmetry [29], is not load-bearing: the paper states 'The results reported in this paper were produced by a lightweight numerical extraction pipeline, developed for this work' (Sec. 5.1). The manuscript itself flags the genuine limitations: in Section 6.5 the centroid-based star 'is not shown to be a consistent discretization' and 'whether ∆_k approximates the continuum screened Poisson operator on this non-well-centered mesh is the open convergence question of Section 7.' The skeptic's orientation-sign objection to Hodge-star equivariance for orientation-reversing isometries, if valid, would be a correctness/falsification concern about Theorem 4.3, not a circular reduction of a prediction to its inputs. No circular step meeting the evidentiary bar is present.

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

No fitted parameters enter the central derivation; the framework rests on standard representation theory and geometric hypotheses about exact mesh symmetry and Hodge-star invertibility. The 3D extension introduces a non-standard centroid star whose consistency is unproven, the main caveat.

free parameters (2)
  • convergence exponent α = fitted in Fig. 8 legend via ρ_∞ − ρ ∝ N^{-α}
    Descriptive fit of approach to the T_d sequential speedup asymptote in Experiment 4; not used in any derivation and not load-bearing.
  • regularization ε = 10^{-6}
    Hand-chosen shift (∆_0+εI) to lift the constant null mode on closed manifolds in the Poisson experiments; standard practice, not load-bearing for equivariance or block-diagonalization.
assumptions (6)
  • standard math Schur's lemma and isotypic decomposition (Eq. 2-4, Theorem 2.1, Lemma 2.2)
    Used in Theorem 5.1 to conclude that an equivariant operator is block-diagonal in the symmetry-adapted basis. Standard finite-group representation theory.
  • standard math Character projection formula Π_j = (n_j/|G|) Σ χ_j(g^{-1}) ρ(g) (Eq. 6)
    Used to construct isotypic projectors and the symmetry-adapted basis (Sections 2.1, 5.1).
  • domain assumption The mesh K carries a finite group G acting by simplicial isomorphisms that are isometries of R^N
    Hypothesis of Theorems 4.1 and 4.3; equivariance of ⋆ requires metric preservation. Approximate symmetry is handled by orbit-averaging (Remark 4.6), which alters the operator.
  • domain assumption The chosen Hodge star is invertible on each cochain degree
    Lemma 4.7 and Theorem 5.1 require ⋆^{-1}; in 2D this follows from strict Delaunay; in 3D from the centroid-based star positivity (Section 6.5), but the centroid star is non-standard.
  • domain assumption The dual cell assignment is G-equivariant: g(⋆σ)=⋆(gσ) for circumcentric and barycentric duals
    Critical for Theorem 4.3(i) and Corollary 4.4; holds because isometries map circumcenters/barycenters to corresponding centers.
  • domain assumption Strict Delaunay / non-degenerate geometry on the 2D test meshes (hexagonal lattice avoids cotangent zeros)
    Needed for positivity and invertibility of ⋆_1 (Section 6.2). The paper excludes the degenerate cocircular case explicitly.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Parallelisation of Discrete Exterior Calculus via Representation Theory on Curved and Three-Dimensional Meshes." pith.science (2026). https://pith.science/paper/CSTJPUYA

@misc{pith2026260725192,
  author       = {Pith},
  title        = {Pith review of: Parallelisation of Discrete Exterior Calculus via Representation Theory on Curved and Three-Dimensional Meshes},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CSTJPUYA}},
  note         = {Machine review of arXiv:2607.25192}
}
abstract

We establish a universal block-diagonalization framework for Discrete Exterior Calculus (DEC) operators on symmetric meshes, enabling embarrassingly parallel solvers with provable FLOP reductions. We prove that the two fundamental DEC operators, the discrete exterior derivative $d$ and the Hodge star $\star$, are equivariant under isometric finite group actions on simplicial complexes. The proof exploits the permutation representation induced on cochain spaces by the group action. As a consequence, any operator assembled from $d$ and $\star$ (including the Hodge Laplacian, the codifferential, Maxwell-type operators, and elasticity operators) inherits a block-diagonal structure in a single symmetry-adapted basis, which is computed only once per mesh. Unlike spectral methods restricted to flat Platonic domains, the framework applies natively to curved manifolds and is applicable in principle to computational electromagnetism and geometric fluid simulation on symmetric domains. Numerical experiments on a geodesic sphere ($I_h$ symmetry) and a hexagonal torus ($D_{6h}$ symmetry) yield FLOP-based parallel speedups, relative to a dense direct factorization, of up to $62\times$ and $182\times$, respectively. A further experiment on a body-centred-cubic (BCC) tessellation of the flat 3-torus $T^3$ with $T_d$ symmetry confirms equivariance of the exterior derivative, Hodge star, and Hodge Laplacian at machine precision for form degrees $k=0,1,2$ across three mesh resolutions. The FLOP-based sequential speedup approaches its theoretical asymptote of $\approx 9.07\times$, which a standard Schur-multiplicity reduction deepens by a further factor of order $|G|$. These results show that a single symmetry-adapted basis reduces the linear-solve cost of structure-preserving DEC computations on curved and three-dimensional meshes.

Figures

Figures reproduced from arXiv: 2607.25192 by the authors.

Figure 1
Figure 1. Block-diagonalization of the DEC building blocks on two symmetric surfaces. Pan [PITH_FULL_IMAGE:figures/full_fig_p017_1.png] view at source ↗
Figure 2
Figure 2. Classification of the ∆1 eigenvalues by irreducible representation, on the two meshes of [PITH_FULL_IMAGE:figures/full_fig_p018_2.png] view at source ↗
Figure 3
Figure 3. Lowest non-trivial eigenfunction of ∆0 in each irreducible representation of Ih, ren￾dered on the geodesic icosphere (n = 4, N0 = 162). Top row: gerade irreps (invariant under inversion). Bottom row: ungerade irreps (anti-invariant under inversion). The Au irrep has multiplicity zero at this resolution. Colour scale: purple (positive) to orange (negative). Spectral fingerprint of the subduction Dℓ ↓ Ih. The block cl… view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Mass-weighted per-degree power P(ℓ) of the lowest non-trivial eigenfunction of ∆0 in each icosahedral irrep, on the geodesic icosphere (n = 6, N0 = 362). Each panel is annotated with the measured dominant degree ℓ ⋆ , which coincides with the subduction prediction in a…
Figure 5
Figure 5. Figure 5: Computational performance of direct vs. symmetry-adapted solve for ∆ [PITH_FULL_IMAGE:figures/full_fig_p024_5.png]
Figure 6
Figure 6. Figure 6: Measured wall-clock parallel speedup versus the predicted FLOP speedup for the [PITH_FULL_IMAGE:figures/full_fig_p025_6.png]
Figure 7
Figure 7. Figure 7: Block sizes of the isotypic decomposition of ∆ [PITH_FULL_IMAGE:figures/full_fig_p027_7.png]
Figure 8
Figure 8. Figure 8: FLOP speedup ratio ρFLOP for the block-diagonalized sequential solve of ∆0, ∆1, ∆2 on the fan-BCC tessellation of T 3 under Td. The dashed line marks the regular-representation sequential limit ρ∞ seq = |G| 3/ P ρ d 6 ρ ≈ 9.07; the upper bound ρ∞ par = |G| 3/d6 max ≈ 1…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

37 extracted references · 14 canonical work pages

  1. [1]

    Hirani.Discrete Exterior Calculus

    Anil N. Hirani.Discrete Exterior Calculus. PhD thesis, California Institute of Technology,

  2. [2]

    Hirani, Melvin Leok, and Jerrold E

    Mathieu Desbrun, Anil N. Hirani, Melvin Leok, and Jerrold E. Marsden. Discrete exterior calculus.arXiv preprint math/0508341, 2005. DOI: doi:10.48550/arXiv.math/0508341

    arXiv doi:10.48550/arxiv.math/0508341 2005

  3. [3]

    Academic Press, 1998

    Alain Bossavit.Computational Electromagnetism: Variational Formulations, Complemen- tarity, Edge Elements. Academic Press, 1998. doi:10.1016/B978-0-12-118740-8.X5000-8

    work page doi:10.1016/b978-0-12-118740-8.x5000-8 1998

  4. [4]

    Digital geometry processing with discrete exterior calculus

    Keenan Crane, Fernando de Goes, Mathieu Desbrun, and Peter Schr¨ oder. Digital geometry processing with discrete exterior calculus. InACM SIGGRAPH 2013 Courses, SIGGRAPH ’13, New York, NY, USA, 2013. ACM. doi:10.1145/2504435.2504442

    arXiv 2013

  5. [5]

    Stable, circulation-preserving, simplicial fluids.ACM Transactions on Graphics, 26(1):4, 2007

    Sharif Elcott, Yiying Tong, Eva Kanso, Peter Schr¨ oder, and Mathieu Desbrun. Stable, circulation-preserving, simplicial fluids.ACM Transactions on Graphics, 26(1):4, 2007. doi:10.1145/1189762.1189766. 30

    arXiv 2007

  6. [6]

    Mohamed, Anil N

    Mamdouh S. Mohamed, Anil N. Hirani, and Ravi Samtaney. Discrete exterior calculus discretization of incompressible Navier–Stokes equations over surface simplicial meshes. Journal of Computational Physics, 312:175–191, 2016. doi:10.1016/j.jcp.2016.02.028

    doi:10.1016/j.jcp.2016.02.028 2016

  7. [7]

    Mohamed, and Ravi Samtaney

    Pankaj Jagad, Abdullah Abukhwejah, Mamdouh S. Mohamed, and Ravi Samtaney. A primitive variable discrete exterior calculus discretization of incompressible Navier– Stokes equations over surface simplicial meshes.Physics of Fluids, 33(1):017114, 2021. doi:10.1063/5.0035981

    work page doi:10.1063/5.0035981 2021

  8. [8]

    Discrete exterior calculus (DEC) for the surface Navier–Stokes equation

    Ingo Nitschke, Sebastian Reuther, and Axel Voigt. Discrete exterior calculus (DEC) for the surface Navier–Stokes equation. In Dieter Bothe and Arnold Reusken, editors,Transport Processes at Fluidic Interfaces, Advances in Mathematical Fluid Mechanics, pages 177–197. Springer, 2017. doi:10.1007/978-3-319-56602-3 7

    work page doi:10.1007/978-3-319-56602-3 2017

Show all 37 references
  1. [9]

    Mohamed, Anil N

    Mamdouh S. Mohamed, Anil N. Hirani, and Ravi Samtaney. Numerical convergence of discrete exterior calculus on arbitrary surface meshes.International Journal for Computational Methods in Engineering Science and Mechanics, 19(3):194–206, 2018. doi:10.1080/15502287.2018.1446196

    2018

  2. [10]

    Allgower, K

    Eugene L. Allgower, K. B¨ ohmer, K. Georg, and R. Miranda.Exploiting symmetry in applied and numerical analysis, volume 29 ofLectures in Applied Mathematics. American Mathematical Society, Providence, RI, 1997. doi:10.1090/lam/029

    1997 doi

  3. [11]

    Birkh¨ auser, 1992

    Albert F¨ assler and Eduard Stiefel.Group Theoretical Methods and Their Applications. Birkh¨ auser, 1992. ISBN 978-1-4612-2824-0. doi:10.1007/978-1-4612-2824-0

    1992 doi

  4. [12]

    Symmetry, groups, and boundary value problems

    Alain Bossavit. Symmetry, groups, and boundary value problems. A progressive introduc- tion to noncommutative harmonic analysis of partial differential equations in domains with geometrical symmetry.Computer Methods in Applied Mechanics and Engineering, 56(2): 167–215, 1986. d...

    1986 doi

  5. [13]

    Munthe-Kaas

    Krister ˚Ahlander and Hans Z. Munthe-Kaas. Applications of the generalized Fourier transform in numerical linear algebra.BIT Numerical Mathematics, 45(4):819–850, 2005. doi:10.1007/s10543-005-0030-3

    2005 doi

  6. [14]

    Parallelisation of partial differential equations via representation the- ory.SIAM Review (to appear), 2025

    Sheehan Olver. Parallelisation of partial differential equations via representation the- ory.SIAM Review (to appear), 2025. URLhttps://arxiv.org/abs/2411.07365.arXiv preprint arXiv:2411.07365

    2025

  7. [15]

    Arnold, Richard S

    Douglas N. Arnold, Richard S. Falk, and Ragnar Winther. Finite element exterior calculus, homological techniques, and applications.Acta Numerica, 15:1–155, 2006. doi:10.1017/S0962492906210018

    2006 doi

  8. [16]

    Arnold, Richard S

    Douglas N. Arnold, Richard S. Falk, and Ragnar Winther. Finite element exterior calculus: from Hodge theory to numerical stability.Bulletin of the American Mathematical Society, 47(2):281–354, 2010. doi:10.1090/S0273-0979-10-01278-4

    2010 doi

  9. [18]

    Martin W. Licht. Symmetry and invariant bases in finite element exterior calculus.Foun- dations of Computational Mathematics, 24(4):1185–1224, 2024. doi:10.1007/s10208-024- 09650-4

    2024 doi

  10. [19]

    Marsden and Matthew West

    Jerrold E. Marsden and Matthew West. Discrete mechanics and variational integrators. Acta Numerica, 10:357–514, 2001. doi:10.1017/S096249290100003X. 31

    2001 doi

  11. [20]

    Bronstein, Joan Bruna, Taco Cohen, and Petar Veliˇ ckovi´ c

    Michael M. Bronstein, Joan Bruna, Taco Cohen, and Petar Veliˇ ckovi´ c. Geometric deep learning: Grids, groups, graphs, geodesics, and gauges.arXiv preprint arXiv:2104.13478,

    arXiv

  12. [21]

    Group equivariant convolutional networks

    Taco Cohen and Max Welling. Group equivariant convolutional networks. InInterna- tional Conference on Machine Learning (ICML), pages 2990–2999, 2016. URLhttp: //proceedings.mlr.press/v48/cohen16.html

    2016

  13. [22]

    Boom, Ashley Seepujak, Odysseas Kosmas, Lee Margetts, and Andrey Jivkov

    Pieter D. Boom, Ashley Seepujak, Odysseas Kosmas, Lee Margetts, and Andrey Jivkov. Parallelized discrete exterior calculus for three-dimensional elliptic prob- lems.Computer Physics Communications, 279:108456, 2022. ISSN 0010-4655. doi:10.1016/j.cpc.2022.108456. URLhttps://doi...

    2022

  14. [23]

    Springer-Verlag, 1977

    Jean-Pierre Serre.Linear Representations of Finite Groups. Springer-Verlag, 1977. ISBN 978-1-4684-9458-7. doi:10.1007/978-1-4684-9458-7

    1977 doi

  15. [24]

    Bobenko and Boris A

    Alexander I. Bobenko and Boris A. Springborn. A discrete Laplace-Beltrami opera- tor for simplicial surfaces.Discrete & Computational Geometry, 38(4):740–756, 2007. doi:10.1007/s00454-007-1339-4

    2007 doi

  16. [25]

    Hirani, Damrong Guoy, and Edgar A

    Evan VanderZee, Anil N. Hirani, Damrong Guoy, and Edgar A. Ramos. Well- centered triangulation.SIAM Journal on Scientific Computing, 31(6):4497–4523, 2010. doi:10.1137/090748214

    2010 doi

  17. [26]

    Hirani, Kalyana B

    Anil N. Hirani, Kalyana B. Nakshatrala, and Jehanzeb H. Chaudhry. Numerical method for Darcy flow derived using Discrete Exterior Calculus.International Journal for Computational Methods in Engineering Science and Mechanics, 16(3):151–169, 2015. doi:10.1080/15502287.2014.977500

    2015

  18. [27]

    Bossavit

    A. Bossavit. Boundary value problems with symmetry and their approximation by finite elements.SIAM J. Appl. Math., 53(5):1352–1380, 1993. doi:10.1137/0153064

    1993 doi

  19. [28]

    Guermond, P

    J.-L. Guermond, P. Minev, and J. Shen. An overview of projection methods for incompressible flows.Comput. Methods Appl. Mech. Engrg., 195:6011–6045, 2006. doi:10.1016/j.cma.2005.10.010

    2006 doi

  20. [29]

    da Silva and Marcelo P

    Leon D. da Silva and Marcelo P. Santos. PySymmetry: A Sage/Python Framework for the Symmetry Reduction of LinearG-equivariant Systems.arXiv preprint arXiv:2509.19479,

  21. [30]

    do Carmo.Differential Geometry of Curves and Surfaces

    Manfredo P. do Carmo.Differential Geometry of Curves and Surfaces. Dover Publications, Mineola, NY, 2 edition, 2016. ISBN 978-0-486-80699-0

    2016

  22. [31]

    Discrete Laplace operators: No free lunch

    Max Wardetzky, Saket Mathur, Felix K¨ alberer, and Eitan Grinspun. Discrete Laplace operators: No free lunch. InProceedings of the Fifth Eurographics Symposium on Geometry Processing (SGP 2007), pages 33–37, 2007. doi:10.2312/SGP/SGP07/033-037

    2007 doi

  23. [32]

    Dresselhaus, Gene Dresselhaus, and Ado Jorio.Group Theory: Application to the Physics of Condensed Matter

    Mildred S. Dresselhaus, Gene Dresselhaus, and Ado Jorio.Group Theory: Application to the Physics of Condensed Matter. Springer, Berlin, Heidelberg, 2008. doi:10.1007/978-3- 540-32899-5

    2008 doi

  24. [33]

    N. V. Cohan. The spherical harmonics with the symmetry of the icosahe- dral group.Proceedings of the Cambridge Philosophical Society, 54(1):28–38, 1958. doi:10.1017/S0305004100033156. 32

    1958 doi

  25. [34]

    Convergence of discrete exterior calculus approx- imations for poisson problems.Discrete & Computational Geometry, 63(2):346–376, 2020

    Holger Schulz and Gantumur Tsogtgerel. Convergence of discrete exterior calculus approx- imations for poisson problems.Discrete & Computational Geometry, 63(2):346–376, 2020. doi:10.1007/s00454-019-00159-x

    2020 doi

  26. [35]

    A framework for analysis of DEC approximations to Hodge-Laplacian problems using generalized Whitney forms, 2025

    Johnny Guzm´ an and Pratyush Potu. A framework for analysis of DEC approximations to Hodge-Laplacian problems using generalized Whitney forms, 2025. URLhttps://arxiv. org/abs/2505.08934.arXiv preprint arXiv:2505.08934. 33

    2025 arXiv

  27. [2003]

    DOI: doi:10.7907/ZHY8-V329

    doi

  28. [2021]

    doi:10.48550/arXiv.2104.13478

    doi arXiv

  29. [2025]

    URLhttps://doi.org/10.48550/arXiv.2509

    doi:10.48550/arXiv.2509.19479. URLhttps://doi.org/10.48550/arXiv.2509. 19479

    doi

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.