arxiv: 2605.11012 · v1 · submitted 2026-05-10 · ❄️ cond-mat.soft

Recognition: 2 theorem links

· Lean Theorem

Inverse Design of Metainterfaces for Static Friction Control: Beyond the Hertzian Limit

Arnav Singhal, Ga\"etan Cortes, Jacopo Bilotto, Jean-Fran\c{c}ois Molinari, Joaquin Garcia-Suarez, Lucas Fourel

Authors on Pith no claims yet

Pith reviewed 2026-05-13 01:59 UTC · model grok-4.3

classification ❄️ cond-mat.soft

keywords inverse designmetainterfacesstatic frictionaxisymmetric asperitiesdifferentiable optimizationcontact mechanicstribologyboundary element method

0 comments

The pith

General axisymmetric asperities enable nonlinear static friction responses beyond standard Hertzian limits through inverse design.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a method to inversely design metainterfaces that achieve programmable static friction by using general axisymmetric asperities instead of classical Hertzian shapes. Standard rough surfaces produce linear area-load scaling that restricts functional range, but this framework discovers non-standard topographies matching complex target friction laws. It embeds a fully differentiable contact mechanics engine inside a neural network and quadratic optimizer, applies regularized physical gradients for automatic discovery, and validates results with high-fidelity boundary element simulations. Applications in soft robotics, haptics, and precision gripping would benefit if custom friction behaviors become routinely achievable.

Core claim

By utilizing general axisymmetric asperities, nonlinear macroscopic responses unattainable by standard Hertzian contacts are unlocked. A fully differentiable contact mechanics engine is embedded within a neural network and a quadratic optimizer, and regularized physical gradients are leveraged to automatically discover non-standard topographies that reproduce complex target friction laws with only a few asperities in unit cells. The predicted designs are strictly validated against high-fidelity Boundary Element Method simulations.

What carries the argument

The fully differentiable contact mechanics engine embedded in a neural network with quadratic optimizer, which supplies regularized physical gradients to guide discovery of asperity topographies that match desired friction laws.

If this is right

Nonlinear friction laws become accessible using metainterfaces with only a few asperities per unit cell.
Data-driven optimization combines with physics-based contact models to generate functional tribological surfaces.
The method provides a scale-invariant route to designing surfaces for specified static friction behaviors.
Validation against boundary element simulations supports reliable prediction of macroscopic contact responses.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same differentiable engine could be adapted to optimize for related properties such as adhesion or electrical contact resistance.
Real fabrication would require checking whether the discovered shapes can be produced at the target scale without introducing unintended defects.
Extending the model to include surface compliance or material nonlinearity would test robustness for broader material classes.

Load-bearing premise

The differentiable contact engine must accurately capture the physics of general axisymmetric asperities at relevant scales, and the gradients must steer the optimizer to designs that remain physically stable and fabricable.

What would settle it

Fabricate prototypes of the discovered asperity shapes and measure their measured friction force versus applied load to determine whether the response matches the target nonlinear law and the boundary element simulation output.

Figures

Figures reproduced from arXiv: 2605.11012 by Arnav Singhal, Ga\"etan Cortes, Jacopo Bilotto, Jean-Fran\c{c}ois Molinari, Joaquin Garcia-Suarez, Lucas Fourel.

**Figure 2.** Figure 2: The physics-informed inverse modeling pipeline based on a deep neural network. The input [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗

**Figure 3.** Figure 3: Topographical test-set sub-domains and corresponding neural reconstructions of contact [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗

**Figure 4.** Figure 4: Training dynamics and hyperparameter schedules for the neural surrogate. (a) Convergence [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗

**Figure 5.** Figure 5: The left panels, (a) and (c), illustrate the macroscopic constitutive response, contact area [PITH_FULL_IMAGE:figures/full_fig_p010_5.png] view at source ↗

**Figure 6.** Figure 6: Inverse design performance on out-of-distribution targets: a saturating limit, a bilinear [PITH_FULL_IMAGE:figures/full_fig_p011_6.png] view at source ↗

**Figure 7.** Figure 7: (Left) The macroscopic contact response, [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗

**Figure 8.** Figure 8: Additional examples of BEM validation plots for representative surfaces from the testing [PITH_FULL_IMAGE:figures/full_fig_p019_8.png] view at source ↗

read the original abstract

Programming the static friction of mechanical interfaces is critical for soft robotics, haptics, and precision gripping. Static friction is governed by the real contact area, and standard rough surfaces exhibit a linear area-load scaling inherent to classical Archard and Greenwood-Williamson models, severely restricting their functional range. Here, we propose a framework for the inverse design of tribological metainterfaces engineered for programmable contact behaviors. By utilizing general axisymmetric asperities, we unlock nonlinear macroscopic responses unattainable by standard Hertzian contacts. To solve the inverse problem, we embed a fully differentiable contact mechanics engine within a neural network and a quadratic optimizer. We leverage regularized physical gradients to automatically discover non-standard topographies that reproduce complex target friction laws, with only a few asperities in unit cells. The predicted designs are strictly validated against high-fidelity Boundary Element Method (BEM) simulations. This framework bridges data-driven optimization and rigorous physics, offering a scale-invariant pathway for discovering functional tribological surfaces.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper gives a workable inverse-design pipeline for non-Hertzian asperities that target arbitrary friction laws, but the validation numbers and regularization checks are missing so the central claim is still unproven.

read the letter

The punchline is that this work shows how to optimize a handful of axisymmetric asperity profiles inside a unit cell so the macroscopic friction curve matches a user-specified nonlinear target. That step is new relative to the classical Archard or Greenwood-Williamson assumptions the abstract cites, and the choice to embed a differentiable contact engine inside a neural net plus quadratic optimizer is a clean way to do the inverse problem without hand-crafted parametrizations. The abstract also states that the resulting shapes were checked in high-fidelity BEM, which is the right sanity test. Those pieces are worth noting. The soft spot is exactly the one the stress-test flags. The abstract mentions regularized physical gradients but gives no error bars on how much the differentiable engine deviates from unsmoothed BEM on the final designs, no ablation on regularization strength, and no check that small perturbations to the discovered topography still recover the target law. For non-Hertzian profiles the contact edges are sharp, so any smoothing artifact would show up most clearly there. Without those numbers the claim that the designs are “strictly validated” rests on an unshown implementation detail. The rest of the mechanics looks standard and the optimization is not circular in the way the reader worried. This paper is for tribologists and soft-robotics groups who already run contact simulations and want a design tool rather than a new physical law. It is coherent enough on its own terms to deserve a serious referee who can ask for the missing quantitative comparisons and regularization study. I would send it to review.

Referee Report

2 major / 2 minor

Summary. The paper claims to introduce an inverse-design framework for tribological metainterfaces that uses general axisymmetric asperities (beyond Hertzian spheres) to achieve user-specified nonlinear static-friction laws. A fully differentiable contact-mechanics engine is embedded in a neural network plus quadratic optimizer; regularized physical gradients are used to discover sparse unit-cell topographies that match target friction curves, with the resulting designs asserted to be strictly validated by high-fidelity BEM simulations.

Significance. If the central claim holds, the work would provide a practical, scale-invariant route to programming macroscopic friction responses that are inaccessible to classical rough-surface models, with direct relevance to soft robotics and precision gripping. The combination of differentiable physics with optimization is a methodological strength that could be extended to other contact problems.

major comments (2)

[Abstract] Abstract: the assertion that the discovered designs are “strictly validated against high-fidelity BEM simulations” is not accompanied by any quantitative discrepancy metrics, L2 errors, or load-area curve comparisons between the regularized differentiable engine and unsmoothed BEM on the final geometries. This information is load-bearing for the claim that the optimizer has found physically realizable non-Hertzian topographies rather than regularization artifacts.
[Methods (differentiable engine)] Methods section describing the differentiable contact engine: no value or functional form is given for the regularization strength applied to the gap/contact condition or load distribution, nor is any ablation or sensitivity study reported. Without this, it is impossible to assess whether the gradients remain faithful to the underlying elastic problem for the sharper contact boundaries that characterize the non-Hertzian profiles highlighted as the paper’s novelty.

minor comments (2)

[Optimization formulation] The precise mathematical definition of the target friction laws (e.g., how the desired area-load curve is encoded as an objective) should be stated explicitly, including whether any of the targets were themselves generated by the same contact model family.
[Figures] Figure captions and axis labels should indicate the number of asperities per unit cell and the range of loads over which the friction curves are compared.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the thoughtful and constructive report. The two major comments identify genuine opportunities to strengthen the quantitative validation and reproducibility of our differentiable contact engine. We address each point below and will revise the manuscript to incorporate the requested details and additional analyses.

read point-by-point responses

Referee: [Abstract] Abstract: the assertion that the discovered designs are “strictly validated against high-fidelity BEM simulations” is not accompanied by any quantitative discrepancy metrics, L2 errors, or load-area curve comparisons between the regularized differentiable engine and unsmoothed BEM on the final geometries. This information is load-bearing for the claim that the optimizer has found physically realizable non-Hertzian topographies rather than regularization artifacts.

Authors: We agree that explicit quantitative metrics are necessary to substantiate the validation claim. The current manuscript presents visual overlays and qualitative agreement between the differentiable predictions and BEM results for the optimized asperity shapes, but does not report L2 norms, relative errors, or direct load-area curve comparisons on the final geometries. In the revised version we will add a dedicated validation subsection (and accompanying figure) that quantifies the discrepancy between the regularized differentiable engine and unsmoothed BEM on the discovered non-Hertzian profiles, including L2 errors on both gap and pressure fields as well as integrated load-area curves. This will directly address whether the optimizer has recovered physically realizable topographies. revision: yes
Referee: [Methods (differentiable engine)] Methods section describing the differentiable contact engine: no value or functional form is given for the regularization strength applied to the gap/contact condition or load distribution, nor is any ablation or sensitivity study reported. Without this, it is impossible to assess whether the gradients remain faithful to the underlying elastic problem for the sharper contact boundaries that characterize the non-Hertzian profiles highlighted as the paper’s novelty.

Authors: We concur that the regularization parameters must be stated explicitly and their influence examined. The differentiable engine employs a smoothed penalty on the gap function and a quadratic regularization on the pressure distribution; the strength parameter was selected via preliminary convergence tests to maintain gradient fidelity while enabling back-propagation. In the revised Methods section we will provide the exact functional form of the regularizer together with the numerical value(s) used throughout the optimization. We will also add a short sensitivity/ablation study demonstrating that the discovered non-Hertzian asperity features and the resulting friction-area relations remain stable across a range of regularization strengths, thereby confirming that the gradients stay faithful to the underlying elastic problem even for the sharper contact edges. revision: yes

Circularity Check

0 steps flagged

No significant circularity; optimization uses external validation

full rationale

The paper presents an inverse-design optimization loop that embeds a differentiable contact engine inside a neural network plus quadratic optimizer to match user-specified target friction laws via regularized gradients. Designs are then validated against an independent high-fidelity BEM solver. No quoted equation or step reduces a claimed prediction to a fitted parameter or self-citation by construction; the targets are externally supplied and the final check is performed outside the differentiable model. Self-citations, if present, are not load-bearing for the central claim.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The framework rests on the assumption that contact mechanics for axisymmetric asperities admits a differentiable formulation suitable for gradient-based optimization; no new physical constants or entities are introduced beyond standard models.

axioms (1)

domain assumption Contact mechanics for general axisymmetric asperities can be formulated as a fully differentiable engine
Invoked when embedding the engine inside the neural network and optimizer

pith-pipeline@v0.9.0 · 5494 in / 1369 out tokens · 34385 ms · 2026-05-13T01:59:46.386141+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear
embed a fully differentiable contact mechanics engine... regularized physical gradients... softplus approximation ⟨x⟩+ ≈ (1/κ)ln(1+exp(κx))... Sneddon forward model
IndisputableMonolith/Foundation/AlphaCoordinateFixation.lean J_uniquely_calibrated_via_higher_derivative unclear
A∼F^{2/(γ+1)}... multi-asperity superposition... BEM validation <3% error

Reference graph

Works this paper leans on

38 extracted references · 38 canonical work pages · 1 internal anchor

[1]

Introduction, in: Allgower, E.L., Georg, K

Allgower, E.L., Georg, K., 1990. Introduction, in: Allgower, E.L., Georg, K. (Eds.), Numerical Continuation Methods: An Introduction. Springer, Berlin, Heidelberg, pp. 1–6. URL:https: //doi.org/10.1007/978-3-642-61257-2_1, doi:10.1007/978-3-642-61257-2_1

work page doi:10.1007/978-3-642-61257-2_1 1990
[2]

Elastic deformation and the laws of friction

Archard, J.F., Allibone, T.E., 1997. Elastic deformation and the laws of friction. Pro- ceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences 243, 190–205. URL:https://royalsocietypublishing.org/doi/10.1098/rspa.1957.0214, doi:10. 1098/rspa.1957.0214

work page doi:10.1098/rspa.1957.0214 1997
[3]

End-to-End Differentiable Physics for Learning and Control, in: Advances in Neural Information Processing Systems, Curran Associates, Inc

de Avila Belbute-Peres, F., Smith, K., Allen, K., Tenenbaum, J., Kolter, J.Z., 2018. End-to-End Differentiable Physics for Learning and Control, in: Advances in Neural Information Processing Systems, Curran Associates, Inc. URL:https://papers.nips.cc/paper_files/paper/2018/ hash/842424a1d0595b76ec4fa03c46e8d755-Abstract.html

work page 2018
[4]

Designing metainterfaces with specified friction laws

Aymard, A., Delplanque, E., Dalmas, D., Scheibert, J., 2024. Designing metainterfaces with specified friction laws. Science 383, 200–204. URL:https://www.science.org/doi/10.1126/ science.adk4234, doi:10.1126/science.adk4234

work page doi:10.1126/science.adk4234 2024
[5]

Curriculum learning, in: Proceedings of the 26th Annual International Conference on Machine Learning, Association for Computing Machinery, New York, NY, USA

Bengio, Y., Louradour, J., Collobert, R., Weston, J., 2009. Curriculum learning, in: Proceedings of the 26th Annual International Conference on Machine Learning, Association for Computing Machinery, New York, NY, USA. pp. 41–48. URL:https://dl.acm.org/doi/10.1145/1553374. 1553380, doi:10.1145/1553374.1553380

work page doi:10.1145/1553374 2009
[6]

Bessa, M.A., Bostanabad, R., Liu, Z., Hu, A., Apley, D.W., Brinson, C., Chen, W., Liu, W.K.,

work page
[8]

Fluid-mediated impact of soft solids

Bilotto, J., Kolinski, J.M., Lecampion, B., Molinari, J.F., Subhash, G., Garcia-Suarez, J., 2024. Fluid-mediated impact of soft solids. Journal of Fluid Mechanics 997, A35. URL:https://www.cambridge.org/core/journals/journal-of-fluid-mechanics/article/ fluidmediated-impact-of-soft-solids/E208CFA7E3181F48FB66DCE0DF20A182, doi:10.1017/ jfm.2024.820

work page 2024
[9]

Pattern Recog- nition153, 110500 (2024).https://doi.org/https://doi.org/10.1016/j

Bonari, J., Paggi, M., Dini, D., 2022. A new finite element paradigm to solve contact prob- lems with roughness. International Journal of Solids and Structures 253, 111643. URL: https://www.sciencedirect.com/science/article/pii/S0020768322001640, doi:10.1016/j. ijsolstr.2022.111643

work page doi:10.1016/j 2022
[10]

Application des potentiels à l’étude de l’équilibre et du mouve- ment des solides élastiques

Boussinesq, J., 1885. Application des potentiels à l’étude de l’équilibre et du mouve- ment des solides élastiques. Gauthier-Villars. URL:https://gallica.bnf.fr/ark:/12148/ bpt6k9651115r

work page
[11]

12 Black-hole perturbation theory,

Bowden, F.P., Tabor, D., 2001. Area of Contact Between Solids, in: Bowden, F.P., Tabor, D. (Eds.), The Friction and Lubrication of Solids. Oxford University Press, p. 0. URL:https:// doi.org/10.1093/oso/9780198507772.003.0002, doi:10.1093/oso/9780198507772.003.0002. 13

work page doi:10.1093/oso/9780198507772.003.0002 2001
[12]

The role of adhesion in contact mechanics

Ciavarella, M., Joe, J., Papangelo, A., Barber, J.R., 2019. The role of adhesion in contact mechanics. Journal of The Royal Society Interface 16, 20180738. URL:https://doi.org/10. 1098/rsif.2018.0738, doi:10.1098/rsif.2018.0738

work page doi:10.1098/rsif.2018.0738 2019
[13]

Inverse design and flexible parameterization of meta-optics using algorithmic differentiation

Colburn, S., Majumdar, A., 2021. Inverse design and flexible parameterization of meta-optics using algorithmic differentiation. Communications Physics 4, 65. URL:https://www.nature. com/articles/s42005-021-00568-6, doi:10.1038/s42005-021-00568-6

work page doi:10.1038/s42005-021-00568-6 2021
[14]

Squeaking at soft–rigid frictional interfaces

Djellouli, A., Albertini, G., Wilt, J., Tournat, V., Weitz, D., Rubinstein, S., Bertoldi, K., 2026. Squeaking at soft–rigid frictional interfaces. Nature 650, 891–897. URL:https://www.nature. com/articles/s41586-026-10132-3, doi:10.1038/s41586-026-10132-3

work page doi:10.1038/s41586-026-10132-3 2026
[15]

Tamaas: a library for elastic-plastic contact of periodic rough surfaces

Frérot, L., Anciaux, G., Rey, V., Pham-Ba, S., Molinari, J.F., 2020. Tamaas: a library for elastic-plastic contact of periodic rough surfaces. Journal of Open Source Software 5, 2121. URL: https://joss.theoj.org/papers/10.21105/joss.02121, doi:10.21105/joss.02121

work page doi:10.21105/joss.02121 2020
[16]

Elastische Beanspruchung des Erdbodens unter Fundamenten

Föppl, L., 1941. Elastische Beanspruchung des Erdbodens unter Fundamenten. Forschung auf dem Gebiet des Ingenieurwesens A 12, 31–39. URL:https://doi.org/10.1007/BF02593958, doi:10.1007/BF02593958

work page doi:10.1007/bf02593958 1941
[17]

A Matter of Shape: Contact Area Optimization in Soft Lubricated Impact

Garcia-Suarez, J., 2026. A Matter of Shape: Contact Area Optimization in Soft Lubricated Impact. Tribology Letters 74, 15. URL:https://doi.org/10.1007/s11249-026-02108-1, doi:10.1007/s11249-026-02108-1

work page doi:10.1007/s11249-026-02108-1 2026
[18]

Contact of nominally flat surfaces

Greenwood, J.A., Williamson, J.B.P., Bowden, F.P., 1997. Contact of nominally flat surfaces. Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences 295, 300–319. URL:https://royalsocietypublishing.org/doi/10.1098/rspa.1966.0242, doi:10. 1098/rspa.1966.0242

work page doi:10.1098/rspa.1966.0242 1997
[19]

Elastic-Plastic Contact Analysis of a Sphere and a Rigid Flat

Kogut, L., Etsion, I., 2002. Elastic-Plastic Contact Analysis of a Sphere and a Rigid Flat. Journal of Applied Mechanics 69, 657–662. URL:https://doi.org/10.1115/1.1490373, doi:10.1115/ 1.1490373

work page doi:10.1115/1.1490373 2002
[20]

Inverse-designed spinodoid meta- materials

Kumar, S., Tan, S., Zheng, L., Kochmann, D.M., 2020. Inverse-designed spinodoid meta- materials. npj Computational Materials 6, 1–10. URL:https://www.nature.com/articles/ s41524-020-0341-6, doi:10.1038/s41524-020-0341-6

work page doi:10.1038/s41524-020-0341-6 2020
[21]

Training Deep Neural Networks for the Inverse Design of Nanophotonic Structures

Liu, D., Tan, Y., Khoram, E., Yu, Z., 2018. Training Deep Neural Networks for the Inverse Design of Nanophotonic Structures. ACS Photonics 5, 1365–1369. URL:https://doi.org/10.1021/ acsphotonics.7b01377, doi:10.1021/acsphotonics.7b01377

work page doi:10.1021/acsphotonics.7b01377 2018
[22]

Decoupled Weight Decay Regularization

Loshchilov, I., Hutter, F., 2019. Decoupled Weight Decay Regularization. URL:http://arxiv. org/abs/1711.05101, doi:10.48550/arXiv.1711.05101. arXiv:1711.05101 [cs]

work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1711.05101 2019
[23]

A Review of Advances in Fabrication Methods and Assistive Technologies of Micro-Structured Surfaces

Ma, Y., Zhang, G., Cao, S., Huo, Z., Han, J., Ma, S., Huang, Z., 2023. A Review of Advances in Fabrication Methods and Assistive Technologies of Micro-Structured Surfaces. Processes 11,

work page 2023
[24]

URL:https://www.mdpi.com/2227-9717/11/5/1337, doi:10.3390/pr11051337

work page doi:10.3390/pr11051337
[25]

A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code

McKay, M.D., Beckman, R.J., Conover, W.J., 1979. A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code. Technometrics 21, 239–245. URL:https://www.jstor.org/stable/1268522, doi:10.2307/1268522

work page doi:10.2307/1268522 1979
[26]

Friction on demand: A generative framework for the inverse design of metainterfaces

Mouton, V., Mélot, A., 2026. Friction on demand: A generative framework for the inverse design of metainterfaces. Tribology International 218, 111724. URL:https://www.sciencedirect.com/ science/article/pii/S0301679X26000666, doi:10.1016/j.triboint.2026.111724

work page doi:10.1016/j.triboint.2026.111724 2026
[27]

Tailoring Frictional Properties of Surfaces Using Diffusion Models

Nordhagen, E., Sveinsson, H.A., Malthe-Sørenssen, A., 2025. Tailoring Frictional Properties of Surfaces Using Diffusion Models. The Journal of Physical Chemistry C 129, 14559–14564. URL: https://pubs.acs.org/doi/10.1021/acs.jpcc.5c02768, doi:10.1021/acs.jpcc.5c02768. 14

work page doi:10.1021/acs.jpcc.5c02768 2025
[28]

Handbook of Contact Mechanics: Exact Solutions of Axisymmetric Contact Problems

Popov, V.L., Heß, M., Willert, E., 2019. Handbook of Contact Mechanics: Exact Solutions of Axisymmetric Contact Problems. Springer Berlin Heidelberg, Berlin, Heidelberg. URL:http: //link.springer.com/10.1007/978-3-662-58709-6, doi:10.1007/978-3-662-58709-6

work page doi:10.1007/978-3-662-58709-6 2019
[29]

Frictional Contact of Soft Polymeric Shells

Sahli, R., Mikkelsen, J., Boye, M.S., Dias, M.A., Aghababaei, R., 2024. Frictional Contact of Soft Polymeric Shells. Physical Review Letters 133, 106202. URL:https://link.aps.org/doi/10. 1103/PhysRevLett.133.106202, doi:10.1103/PhysRevLett.133.106202

work page doi:10.1103/physrevlett.133.106202 2024
[30]

Science Advances , author =

Sanner, A., Kumar, N., Dhinojwala, A., Jacobs, T.D.B., Pastewka, L., 2024. Why soft contacts are stickier when breaking than when making them. Science Advances 10, eadl1277. URL:https: //www.science.org/doi/full/10.1126/sciadv.adl1277, doi:10.1126/sciadv.adl1277

work page doi:10.1126/sciadv.adl1277 2024
[31]

Zur Frage der Druckverteilung unter elastisch gelagerten Tragwerken

Schubert, G., 1942. Zur Frage der Druckverteilung unter elastisch gelagerten Tragwerken. Ingenieur-Archiv 13, 132–147. URL:https://doi.org/10.1007/BF02095912, doi:10.1007/ BF02095912

work page doi:10.1007/bf02095912 1942
[32]

Real Area of Contact in a Soft Transparent Interface by Particle Exclusion Microscopy

Schulze, K.D., Bennett, A.I., Marshall, S., Rowe, K.G., Dunn, A.C., 2016. Real Area of Contact in a Soft Transparent Interface by Particle Exclusion Microscopy. Journal of Tribology 138. URL: https://doi.org/10.1115/1.4032822, doi:10.1115/1.4032822

work page doi:10.1115/1.4032822 2016
[33]

The relation between load and penetration in the axisymmetric boussinesq problem for a punch of arbitrary profile

Sneddon, I.N., 1965. The relation between load and penetration in the axisymmetric boussinesq problem for a punch of arbitrary profile. International Journal of Engineering Science 3, 47–

work page 1965
[34]

1016/0020-7225(65)90019-4

URL:https://www.sciencedirect.com/science/article/pii/0020722565900194, doi:10. 1016/0020-7225(65)90019-4

work page arXiv
[35]

The nonlinear nature of friction

Urbakh, M., Klafter, J., Gourdon, D., Israelachvili, J., 2004. The nonlinear nature of friction. Nature 430, 525–528. URL:https://www.nature.com/articles/nature02750, doi:10.1038/ nature02750

work page 2004
[36]

Persson’s theory of purely normal elastic rough surface contact: A tutorial based on stochastic process theory

Xu, Y., Li, X., Chen, Q., Zhou, Y., 2024. Persson’s theory of purely normal elastic rough surface contact: A tutorial based on stochastic process theory. International Journal of Solids and Structures 290, 112684. URL:https://www.sciencedirect.com/science/article/pii/ S0020768324000416, doi:10.1016/j.ijsolstr.2024.112684

work page doi:10.1016/j.ijsolstr.2024.112684 2024
[37]

JAX-FEM: A differen- tiable GPU-accelerated 3D finite element solver for automatic inverse design and mechanistic data science

Xue, T., Liao, S., Gan, Z., Park, C., Xie, X., Liu, W.K., Cao, J., 2023. JAX-FEM: A differen- tiable GPU-accelerated 3D finite element solver for automatic inverse design and mechanistic data science. Computer Physics Communications 291, 108802. URL:https://www.sciencedirect. com/science/article/pii/S0010465523001479, doi:10.1016/j.cpc.2023.108802

work page doi:10.1016/j.cpc.2023.108802 2023
[38]

The Contact of Elastic Regular Wavy Surfaces Revisited

Yastrebov, V.A., Anciaux, G., Molinari, J.F., 2014. The Contact of Elastic Regular Wavy Surfaces Revisited. TribologyLetters56, 171–183. URL:https://doi.org/10.1007/s11249-014-0395-z, doi:10.1007/s11249-014-0395-z

work page doi:10.1007/s11249-014-0395-z 2014
[39]

Friction Anisotropy with Respect to Topographic Orientation

Yu, C., Wang, Q.J., 2012. Friction Anisotropy with Respect to Topographic Orientation. Scientific Reports 2, 988. URL:https://www.nature.com/articles/srep00988, doi:10.1038/srep00988. A Details on Neural Network Architecture and Training The inverse surrogate model is designed to map macroscopic mechanical responses back to microscopic topography paramete...

work page doi:10.1038/srep00988 2012