pith. sign in

arxiv: 2605.14898 · v1 · pith:4EMK5KKFnew · submitted 2026-05-14 · ❄️ cond-mat.mtrl-sci

Generative reconstruction of 2D and 3D polycrystalline microstructures using symmetrized hyperspherical harmonics

Ali R. Safi , Paul Seibert , Santiago Benito , Alexander Ra{\ss}loff , Markus K\"astner , Benjamin Klusemann This is my paper

Pith reviewed 2026-06-30 20:17 UTC · model grok-4.3

classification ❄️ cond-mat.mtrl-sci
keywords microstructure reconstructionpolycrystalline materialshyperspherical harmonicsquaternionsorientation distributiongenerative modelingspatial correlationsaluminum alloy
0
0 comments X

The pith

A framework reconstructs 3D polycrystalline microstructures from 2D orientation data using quaternion-based symmetrized hyperspherical harmonics.

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

The paper presents a differentiable method that generates three-dimensional voxelized microstructures matching the statistical properties of a target polycrystalline material when given only two-dimensional orientation maps. It replaces Euler-angle representations with unit quaternions and symmetrized hyperspherical harmonics to obtain a continuous, symmetry-invariant description of crystallographic orientations. Reconstructions are obtained by minimizing a loss that combines two-point spatial correlations, a hybrid three-point variogram, and a mean-variation regularizer, using the L-BFGS-B optimizer. The approach is shown to recover both grain morphology and texture in realizations of an aluminum alloy processed by thermo-mechanical treatment.

Core claim

Unit quaternions combined with symmetrized hyperspherical harmonics supply a continuous, symmetry-invariant orientation field that drives descriptor-based reconstruction. The loss is formed from two-point spatial correlations, a novel hybrid three-point variogram, and a mean-variation regularizer. When applied to two-dimensional orientation data of a thermo-mechanically processed aluminum alloy, second-order gradient optimization with L-BFGS-B produces three-dimensional realizations whose morphological features and crystallographic distributions match the reference statistics.

What carries the argument

Symmetrized hyperspherical harmonics on unit quaternions, which furnish a continuous symmetry-invariant orientation descriptor inside an optimization loop that matches two-point correlations, a hybrid three-point variogram, and a mean-variation regularizer.

If this is right

  • Three-dimensional representative volume elements can be synthesized directly from two-dimensional EBSD maps for use in full-field simulations.
  • The combined descriptors preserve both global texture and local interfacial topology.
  • Gradient-based optimization with L-BFGS-B reaches low residuals on the complex loss landscape.
  • An open-source implementation supplies a practical tool for generating microstructure inputs in materials design workflows.

Where Pith is reading between the lines

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

  • The descriptor set may generalize to other alloy systems or processing routes if the same correlation functions remain sufficient.
  • Successful 3D recovery from 2D sections could reduce reliance on serial-sectioning or tomography experiments.
  • The method supplies a route to couple microstructure generation directly with property-prediction pipelines.
  • If the regularizer proves robust, similar differentiable pipelines could be adapted to reconstruct other heterogeneous media such as porous or composite materials.

Load-bearing premise

The chosen two-point correlations, hybrid three-point variogram, and mean-variation regularizer together capture the full statistical ensemble required for faithful reconstruction.

What would settle it

A set of reconstructed volumes that satisfy the reported descriptors yet deviate measurably from independent higher-order spatial statistics or from measured mechanical response of the same aluminum alloy.

Figures

Figures reproduced from arXiv: 2605.14898 by Alexander Ra{\ss}loff, Ali R. Safi, Benjamin Klusemann, Markus K\"astner, Paul Seibert, Santiago Benito.

Figure 1
Figure 1. Figure 1: Comparative visualization of the local crystallographic orientation state across a polycrystalline microstructure [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Spatial distribution of selected SHSH basis function evaluations mapped onto the polycrystalline microstruc [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 1
Figure 1. Figure 1: The resulting fields show distinct orientation regions with physically consistent transitions and clearly indicate [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 3
Figure 3. Figure 3: Grid of auto- and cross-correlation maps for selected SHSH basis functions. The plots visualize the two [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Variogram maps for selected SHSH basis modes. The plots display the first-order variogram [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Schematic workflow of the DMCR framework expanded to orientation information. The pipeline begins [PITH_FULL_IMAGE:figures/full_fig_p011_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Comparison of reconstructed microstructure [PITH_FULL_IMAGE:figures/full_fig_p012_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Influence of the relative weighting between [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: The baseline was held constant to isolate the effect of the regularizer. While other weight combinations may [PITH_FULL_IMAGE:figures/full_fig_p012_8.png] view at source ↗
Figure 8
Figure 8. Figure 8: Influence of the weight wV on a reconstruction with fixed spatial descriptor weights (wS = 1, wγ3 = 10). Increasing regularization progresses from noise suppression (left) to morphological oversmoothing and eventual structural collapse (right) [PITH_FULL_IMAGE:figures/full_fig_p013_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Convergence trajectory of the multi-objective loss function [PITH_FULL_IMAGE:figures/full_fig_p013_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Inverse pole figure (IPF) maps, exemplarily for aluminum microstructures produced via thermo-mechanical [PITH_FULL_IMAGE:figures/full_fig_p014_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Reconstruction results for the statistically homogeneous microstructure, highlighting (a) the original EBSD [PITH_FULL_IMAGE:figures/full_fig_p015_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Comparative error analysis of the 2D orientation variograms for selected SHSH modes ( [PITH_FULL_IMAGE:figures/full_fig_p016_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: 3D realization of an orientation field reconstructed from a single 2D experimental slice, showing (a) the [PITH_FULL_IMAGE:figures/full_fig_p018_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Reconstruction results for the statistically inhomogeneous microstructure, highlighting (a) the original EBSD [PITH_FULL_IMAGE:figures/full_fig_p019_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Evaluation of the SHSH functions for the first nine [PITH_FULL_IMAGE:figures/full_fig_p027_15.png] view at source ↗
Figure 15
Figure 15. Figure 15: (continued) 28 [PITH_FULL_IMAGE:figures/full_fig_p028_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: Two distinct hypothetical microstructural states for interpolation analysis with (a) an isotropic microstructure [PITH_FULL_IMAGE:figures/full_fig_p029_16.png] view at source ↗
Figure 17
Figure 17. Figure 17: Non-linear evolution of microstructural morphology during statistical interpolation after 1000 iterations. The [PITH_FULL_IMAGE:figures/full_fig_p030_17.png] view at source ↗
Figure 18
Figure 18. Figure 18: Comparison of convergence trajectories for various optimization algorithms. L-BFGS-B and Adam show the [PITH_FULL_IMAGE:figures/full_fig_p030_18.png] view at source ↗
Figure 19
Figure 19. Figure 19: Convergence variability for ten independent stochastic realizations. Individual loss lines (grey) show [PITH_FULL_IMAGE:figures/full_fig_p031_19.png] view at source ↗
read the original abstract

Establishing structure-property linkages in polycrystalline materials requires representative two- (2D) and three- (3D) dimensional microstructural inputs for full-field simulations. A core objective of microstructure characterization and reconstruction is the generative synthesis of 2D and 3D microstructures that reflect a target statistical ensemble using limited 2D data as a reference. This work introduces an orientation-based differentiable microstructure characterization and reconstruction framework, implemented in MCRpy, to perform reconstructions of voxelized images. Unit quaternions in combination with symmetrized hyperspherical harmonics are utilized to derive a continuous, symmetry-invariant representation of crystallographic orientations to overcome the numerical singularities and discontinuities associated with traditional Euler-based methods. The descriptor-based reconstructions are driven by a set combining two-point spatial correlations, a novel hybrid three-point variogram, and a mean variation regularizer to capture both global texture and local interfacial topology. The framework's efficiency is demonstrated by reconstructing 3D realizations from 2D orientation data of an aluminum alloy after thermo-mechanical processing, successfully recovering both morphological features and crystallographic distribution. Systematic benchmarking indicates that second-order gradient-based optimization, utilizing the L-BFGS-B algorithm, effectively navigates the complex loss landscape to generate high-fidelity realizations with minimal residuals. This methodology provides a versatile, open-source framework for the digital synthesis of polycrystalline representative volume elements to facilitate the rapid development of microstructure-informed materials design workflows.

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.

Referee Report

2 major / 2 minor

Summary. The paper introduces an orientation-based differentiable microstructure characterization and reconstruction framework implemented in MCRpy. It employs unit quaternions combined with symmetrized hyperspherical harmonics to obtain a continuous, symmetry-invariant representation of crystallographic orientations, avoiding singularities associated with Euler angles. Reconstructions of voxelized 2D and 3D images are driven by a combined descriptor set consisting of two-point spatial correlations, a novel hybrid three-point variogram, and a mean variation regularizer. The framework is demonstrated by generating 3D realizations from 2D orientation data of a thermo-mechanically processed aluminum alloy, with the claim that both morphological features and crystallographic distributions are recovered. Systematic benchmarking shows that L-BFGS-B optimization effectively minimizes the loss to produce high-fidelity realizations. The work provides an open-source tool for synthesizing polycrystalline representative volume elements.

Significance. If the central claims hold, the work offers a meaningful contribution to microstructure reconstruction by supplying a differentiable, symmetry-aware pipeline that directly addresses limitations of Euler-angle representations. The open-source release in MCRpy and the emphasis on gradient-based optimization of a composite loss are concrete strengths that could accelerate microstructure-informed materials design. The hybrid three-point variogram is presented as a targeted addition for interfacial topology, which, if validated, would extend the utility of descriptor-based methods beyond standard two-point statistics.

major comments (2)
  1. [Abstract] Abstract: The claim that the framework 'successfully recover[s] both morphological features and crystallographic distribution' and produces 'high-fidelity realizations with minimal residuals' is presented without any quantitative error metrics (e.g., L2 norms on the two-point correlations or three-point variogram residuals), baseline comparisons to existing reconstruction algorithms, or explicit description of how residuals were computed. This quantitative gap is load-bearing for the central claim of faithful ensemble recovery.
  2. [Results / descriptor section] Results / descriptor section: The assertion that the combination of two-point spatial correlations, the novel hybrid three-point variogram, and the mean variation regularizer is sufficient to determine the full 3D statistical ensemble from 2D data is not accompanied by an ablation study (e.g., reconstruction quality with the three-point term removed) or comparison against independent 3D ground-truth statistics. Without such evidence, it remains unclear whether higher-order spatial correlations or long-range orientation correlations are truly redundant for the aluminum alloy case.
minor comments (2)
  1. [Methods] Notation for the symmetrized hyperspherical harmonics and the precise definition of the hybrid three-point variogram should be given explicitly with equations in the methods section to allow reproducibility.
  2. [Figures] Figure captions should include the specific aluminum alloy composition, processing conditions, and voxel resolution to contextualize the qualitative demonstrations.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for the thoughtful and constructive report. The two major comments highlight opportunities to strengthen the quantitative presentation and validation of our claims. We address each point below and will revise the manuscript accordingly.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The claim that the framework 'successfully recover[s] both morphological features and crystallographic distribution' and produces 'high-fidelity realizations with minimal residuals' is presented without any quantitative error metrics (e.g., L2 norms on the two-point correlations or three-point variogram residuals), baseline comparisons to existing reconstruction algorithms, or explicit description of how residuals were computed. This quantitative gap is load-bearing for the central claim of faithful ensemble recovery.

    Authors: We agree that the abstract would benefit from explicit quantitative metrics. In the revised version we will insert concise statements of the L2 residuals on the two- and three-point descriptors (computed as the Euclidean norm between target and reconstructed correlation functions, normalized by the number of bins) together with a brief description of the residual evaluation. Baseline comparisons to other reconstruction algorithms lie outside the scope of the present methodological contribution but can be noted as future work. revision: yes

  2. Referee: [Results / descriptor section] Results / descriptor section: The assertion that the combination of two-point spatial correlations, the novel hybrid three-point variogram, and the mean variation regularizer is sufficient to determine the full 3D statistical ensemble from 2D data is not accompanied by an ablation study (e.g., reconstruction quality with the three-point term removed) or comparison against independent 3D ground-truth statistics. Without such evidence, it remains unclear whether higher-order spatial correlations or long-range orientation correlations are truly redundant for the aluminum alloy case.

    Authors: We acknowledge that an ablation study would provide additional support for the sufficiency of the chosen descriptor set. We will add a targeted ablation (removing the hybrid three-point term) to the revised results section and report the resulting increase in descriptor mismatch. Independent 3D ground-truth volumes are not available for this thermo-mechanically processed alloy; the reconstruction is performed from 2D sections precisely because only planar data exist. We will clarify this limitation and note that the low residuals achieved with the composite descriptor set constitute the primary evidence of sufficiency for the reported case. revision: partial

standing simulated objections not resolved
  • Independent 3D ground-truth statistics are unavailable for the aluminum alloy specimen, precluding direct comparison to full 3D ensemble measures.

Circularity Check

0 steps flagged

No circularity: independent descriptors and external validation

full rationale

The paper defines new independent components (quaternion + symmetrized hyperspherical harmonics representation, novel hybrid three-point variogram, mean-variation regularizer) and drives L-BFGS-B optimization against external 2D reference data from an aluminum alloy. Reconstructions are evaluated for morphological and crystallographic fidelity on held-out statistics; no equation reduces a claimed prediction to a fitted input by construction, and no load-bearing premise rests on a self-citation chain. The central claim therefore remains falsifiable against independent 3D ground truth.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The approach rests on standard domain assumptions about microstructure statistics being adequately described by low-order spatial correlations; no free parameters or invented entities are explicitly introduced in the abstract.

axioms (2)
  • domain assumption Unit quaternions combined with symmetrized hyperspherical harmonics yield a continuous, symmetry-invariant representation of crystallographic orientations that avoids singularities of Euler angles.
    Invoked to justify the orientation descriptor choice over traditional methods.
  • domain assumption Two-point correlations plus a hybrid three-point variogram plus mean variation regularizer are sufficient to drive faithful generative reconstruction of the target statistical ensemble.
    Central to the descriptor-based optimization objective.

pith-pipeline@v0.9.1-grok · 5812 in / 1470 out tokens · 34067 ms · 2026-06-30T20:17:10.933886+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

79 extracted references · 71 canonical work pages · 3 internal anchors

  1. [1]

    R. A. Lebensohn, C. N. Tomé, A self-consistent anisotropic approach for the simulation of plastic deformation and texture development of polycrystals: Application to zirconium alloys, Acta Metallurgica et Materialia 41 (9) (1993) 2611–2624.doi:10.1016/0956-7151(93)90130-K. URLhttps://www.sciencedirect.com/science/article/pii/095671519390130K

    work page doi:10.1016/0956-7151(93)90130-k 1993

  2. [2]

    Steinbach, F

    I. Steinbach, F. Pezzolla, A generalized field method for multiphase transformations using interface fields, Physica D 134 (4) (1999) 385–393.doi:10.1016/S0167-2789(99)00129-3. URLhttps://linkinghub.elsevier.com/retrieve/pii/S0167278999001293

    work page doi:10.1016/s0167-2789(99)00129-3 1999

  3. [3]

    Chen, Phase-Field Models for Microstructure Evolution, Annu

    L.-Q. Chen, Phase-Field Models for Microstructure Evolution, Annu. Rev. Mater. Res. 32 (V olume 32, 2002) (2002) 113–140, publisher: Annual Reviews.doi:10.1146/annurev.matsci.32.112001.132041. URL https://www.annualreviews.org/content/journals/10.1146/annurev.matsci.32.112001. 132041

    work page doi:10.1146/annurev.matsci.32.112001.132041 2002

  4. [4]

    Roters, M

    F. Roters, M. Diehl, P. Shanthraj, P. Eisenlohr, C. Reuber, S. L. Wong, T. Maiti, A. Ebrahimi, T. Hochrainer, H. O. Fabritius, S. Nikolov, M. Friák, N. Fujita, N. Grilli, K. G. F. Janssens, N. Jia, P. J. J. Kok, D. Ma, F. Meier, E. Werner, M. Stricker, D. Weygand, D. Raabe, DAMASK – The Düsseldorf Advanced Material Simulation Kit for modeling multi-physic...

    work page doi:10.1016/j.commatsci.2018.04.030 2019

  5. [5]

    Yaghoobi, S

    M. Yaghoobi, S. Ganesan, S. Sundar, A. Lakshmanan, S. Rudraraju, J. E. Allison, V . Sundararaghavan, PRISMS- Plasticity: An open-source crystal plasticity finite element software, Computational Materials Science 169 (2019) 109078.doi:10.1016/j.commatsci.2019.109078. URLhttps://www.sciencedirect.com/science/article/pii/S0927025619303696

    work page doi:10.1016/j.commatsci.2019.109078 2019

  6. [6]

    J. M. Hestroffer, J.-C. Stinville, M.-A. Charpagne, M. P. Miller, T. M. Pollock, I. J. Beyerlein, Subsurface microstructure effects on surface resolved slip activity, Journal of the Mechanics and Physics of Solids 196 (2025) 106023.doi:10.1016/j.jmps.2024.106023. URLhttps://www.sciencedirect.com/science/article/pii/S0022509624004897

    work page doi:10.1016/j.jmps.2024.106023 2025

  7. [7]

    M. A. Groeber, B. K. Haley, M. D. Uchic, D. M. Dimiduk, S. Ghosh, 3D reconstruction and characterization of polycrystalline microstructures using a FIB–SEM system, Materials Characterization 57 (4) (2006) 259–273. doi:10.1016/j.matchar.2006.01.019. URLhttps://www.sciencedirect.com/science/article/pii/S1044580306000623 21

    work page doi:10.1016/j.matchar.2006.01.019 2006

  8. [8]

    R. Wirth, Focused Ion Beam (FIB) combined with SEM and TEM: Advanced analytical tools for studies of chemical composition, microstructure and crystal structure in geomaterials on a nanometre scale, Chemical Geology 261 (3) (2009) 217–229.doi:10.1016/j.chemgeo.2008.05.019. URLhttps://www.sciencedirect.com/science/article/pii/S0009254108001940

    work page doi:10.1016/j.chemgeo.2008.05.019 2009

  9. [9]

    Calcagnotto, D

    M. Calcagnotto, D. Ponge, E. Demir, D. Raabe, Orientation gradients and geometrically necessary dislocations in ultrafine grained dual-phase steels studied by 2D and 3D EBSD, Materials Science and Engineering: A 527 (10) (2010) 2738–2746.doi:10.1016/j.msea.2010.01.004. URLhttps://www.sciencedirect.com/science/article/pii/S0921509310000195

    work page doi:10.1016/j.msea.2010.01.004 2010

  10. [10]

    S. F. Li, J. Lind, C. M. Hefferan, R. Pokharel, U. Lienert, A. D. Rollett, R. M. Suter, Three-dimensional plastic response in polycrystalline copper via near-field high-energy X-ray diffraction microscopy, Journal of Applied Crystallography 45 (6) (2012) 1098–1108, publisher: International Union of Crystallography. doi: 10.1107/S0021889812039519. URL//jou...

    work page doi:10.1107/s0021889812039519 2012

  11. [11]

    Gustafson, W

    S. Gustafson, W. Ludwig, P. Shade, D. Naragani, D. Pagan, P. Cook, C. Yildirim, C. Detlefs, M. D. Sangid, Quantifying microscale drivers for fatigue failure via coupled synchrotron X-ray characterization and simu- lations, Nature Communications 11 (1) (2020) 3189, publisher: Nature Publishing Group. doi:10.1038/ s41467-020-16894-2. URLhttps://www.nature.c...

    2020

  12. [12]

    D. C. Pagan, C. R. Pash, A. R. Benson, M. P. Kasemer, Graph neural network modeling of grain-scale anisotropic elastic behavior using simulated and measured microscale data, npj Computational Materials 8 (1) (2022) 259, publisher: Nature Publishing Group.doi:10.1038/s41524-022-00952-y. URLhttps://www.nature.com/articles/s41524-022-00952-y

    work page doi:10.1038/s41524-022-00952-y 2022

  13. [13]

    Yi Wang, J

    W. Yi Wang, J. Li, W. Liu, Z.-K. Liu, Integrated computational materials engineering for advanced materials: A brief review, Computational Materials Science 158 (2019) 42–48. doi:10.1016/j.commatsci.2018.11.001. URLhttps://www.sciencedirect.com/science/article/pii/S0927025618307171

    work page doi:10.1016/j.commatsci.2018.11.001 2019

  14. [14]

    C. L. Y . Yeong, S. Torquato, Reconstructing random media, Physical Review E 57 (1) (1998) 495–506, publisher: American Physical Society.doi:10.1103/PhysRevE.57.495. URLhttps://link.aps.org/doi/10.1103/PhysRevE.57.495

    work page doi:10.1103/physreve.57.495 1998

  15. [15]

    Torquato, Optimal Design of Heterogeneous Materials, Annual Review of Materials Research 40 (1) (2010) 101–129.doi:10.1146/annurev-matsci-070909-104517

    S. Torquato, Optimal Design of Heterogeneous Materials, Annual Review of Materials Research 40 (1) (2010) 101–129.doi:10.1146/annurev-matsci-070909-104517. URLhttps://www.annualreviews.org/doi/10.1146/annurev-matsci-070909-104517

    work page doi:10.1146/annurev-matsci-070909-104517 2010

  16. [16]

    Bostanabad, Y

    R. Bostanabad, Y . Zhang, X. Li, T. Kearney, L. C. Brinson, D. W. Apley, W. K. Liu, W. Chen, Computational microstructure characterization and reconstruction: Review of the state-of-the-art techniques, Progress in Materials Science 95 (2018) 1–41.doi:10.1016/j.pmatsci.2018.01.005. URLhttps://linkinghub.elsevier.com/retrieve/pii/S0079642518300112

    work page doi:10.1016/j.pmatsci.2018.01.005 2018

  17. [17]

    Bargmann, B

    S. Bargmann, B. Klusemann, J. Markmann, J. E. Schnabel, K. Schneider, C. Soyarslan, J. Wilmers, Generation of 3D representative volume elements for heterogeneous materials: A review, Prog. Mater Sci. 96 (2018) 322–384. doi:10.1016/j.pmatsci.2018.02.003. URLhttps://linkinghub.elsevier.com/retrieve/pii/S0079642518300161

    work page doi:10.1016/j.pmatsci.2018.02.003 2018

  18. [18]

    Ghosh, D

    S. Ghosh, D. Dimiduk, D. Furrer, Statistically equivalent representative volume elements (SERVE) for material behaviour analysis and multiscale modelling, International Materials Reviews 68 (8) (2023) 1158–1191, publisher: SAGE Publications.doi:10.1080/09506608.2023.2246766. URLhttps://doi.org/10.1080/09506608.2023.2246766

    work page doi:10.1080/09506608.2023.2246766 2023

  19. [19]

    M. Groeber, A framework for automated analysis and simulation of 3D polycrystalline microstructures.Part 1: Statistical characterization, Acta Materialia 56 (6) (2008) 1257–1273.doi:10.1016/j.actamat.2007.11.041. URLhttps://linkinghub.elsevier.com/retrieve/pii/S1359645407007914

    work page doi:10.1016/j.actamat.2007.11.041 2008

  20. [20]

    M. A. Groeber, M. A. Jackson, DREAM.3D: A Digital Representation Environment for the Analysis of Mi- crostructure in 3D, Integrating Materials and Manufacturing Innovation 3 (1) (2014) 56–72. doi:10.1186/ 2193-9772-3-5. URLhttps://link.springer.com/10.1186/2193-9772-3-5

    work page doi:10.1186/2193-9772-3-5 2014

  21. [21]

    R. Quey, P. Dawson, F. Barbe, Large-scale 3D random polycrystals for the finite element method: Generation, meshing and remeshing, Computer Methods in Applied Mechanics and Engineering 200 (17-20) (2011) 1729– 1745.doi:10.1016/j.cma.2011.01.002. URLhttps://linkinghub.elsevier.com/retrieve/pii/S004578251100003X 22

    work page doi:10.1016/j.cma.2011.01.002 2011

  22. [22]

    Prasad, N

    M. Prasad, N. Vajragupta, A. Hartmaier, Kanapy: A Python package for generating complex synthetic polycrys- talline microstructures, Journal of Open Source Software 4 (43) (2019) 1732.doi:10.21105/joss.01732. URLhttps://joss.theoj.org/papers/10.21105/joss.01732

    work page doi:10.21105/joss.01732 2019

  23. [23]

    Henrich, F

    M. Henrich, F. Pütz, S. Münstermann, M. Henrich, F. Pütz, S. Münstermann, A Novel Approach to Discrete Rep- resentative V olume Element Automation and Generation-DRAGen, Materials 13 (8), company: Multidisciplinary Digital Publishing Institute Distributor: Multidisciplinary Digital Publishing Institute Institution: Multidisciplinary Digital Publishing Ins...

    work page doi:10.3390/ma13081887 2020

  24. [24]

    Henrich, N

    M. Henrich, N. Fehlemann, F. Bexter, M. Neite, L. Kong, F. Shen, M. Könemann, M. Dölz, S. Münstermann, DRAGen – A deep learning supported RVE generator framework for complex microstructure models, Heliyon 9 (8), publisher: Elsevier (Aug. 2023).doi:10.1016/j.heliyon.2023.e19003. URLhttps://www.cell.com/heliyon/abstract/S2405-8440(23)06211-4

    work page doi:10.1016/j.heliyon.2023.e19003 2023

  25. [25]

    Furat, L

    O. Furat, L. Petrich, D. P. Finegan, D. Diercks, F. Usseglio-Viretta, K. Smith, V . Schmidt, Artificial generation of representative single Li-ion electrode particle architectures from microscopy data, npj Computational Materials 7 (1) (2021) 105, publisher: Nature Publishing Group.doi:10.1038/s41524-021-00567-9. URLhttps://www.nature.com/articles/s41524-...

    work page doi:10.1038/s41524-021-00567-9 2021

  26. [26]

    Fuchs, O

    L. Fuchs, O. Furat, D. P. Finegan, J. Allen, F. L. E. Usseglio-Viretta, B. Ozdogru, P. J. Weddle, K. Smith, V . Schmidt, Generating multi-scale Li-ion battery cathode particles with radial grain architectures using stereological generative adversarial networks, Communications Materials 6 (1) (2025) 4, publisher: Nature Publishing Group. doi: 10.1038/s4324...

    work page doi:10.1038/s43246-024-00728-5 2025

  27. [27]

    Yadegari, S

    S. Yadegari, S. Turteltaub, A. S. J. Suiker, P. J. J. Kok, Analysis of banded microstructures in multiphase steels assisted by transformation-induced plasticity, Computational Materials Science 84 (2014) 339–349. doi: 10.1016/j.commatsci.2013.12.002. URLhttps://www.sciencedirect.com/science/article/pii/S0927025613007519

    work page doi:10.1016/j.commatsci.2013.12.002 2014

  28. [28]

    F. J. Humphreys, M. Hatherly, Recrystallization and related annealing phenomena, 1st Edition, Pergamon, Oxford, OX, UK ; Tarrytown, N.Y ., U.S.A, 1995

    1995

  29. [29]

    U. F. Suhuddin, L. Rath, R. M. Halak, B. Klusemann, Microstructure evolution and texture development during production of homogeneous fine-grained aluminum wire by friction extrusion, Materials Characterization 205 (2023) 113252.doi:10.1016/j.matchar.2023.113252. URLhttps://linkinghub.elsevier.com/retrieve/pii/S1044580323006113

    work page doi:10.1016/j.matchar.2023.113252 2023

  30. [30]

    S. A. H. Motaman, F. Roters, C. Haase, Anisotropic polycrystal plasticity due to microstructural heterogeneity: A multi-scale experimental and numerical study on additively manufactured metallic materials, Acta Materialia 185 (2020) 340–369, publisher: Elsevier. URLhttps://www.sciencedirect.com/science/article/pii/S1359645419308298

    2020

  31. [31]

    Sundararaghavan, N

    V . Sundararaghavan, N. Zabaras, Classification and reconstruction of three-dimensional microstructures using support vector machines, Computational Materials Science 32 (2) (2005) 223–239.doi:10.1016/j.commatsci. 2004.07.004. URLhttps://www.sciencedirect.com/science/article/pii/S0927025604001934

    work page doi:10.1016/j.commatsci 2005

  32. [32]

    Kench, S

    S. Kench, S. J. Cooper, Generating three-dimensional structures from a two-dimensional slice with generative adversarial network-based dimensionality expansion, Nature Machine Intelligence 3 (4) (2021) 299–305. doi: 10.1038/s42256-021-00322-1. URLhttps://www.nature.com/articles/s42256-021-00322-1

    work page doi:10.1038/s42256-021-00322-1 2021

  33. [33]

    Zhang, P

    Y . Zhang, P. Seibert, A. Otto, A. Raßloff, M. Ambati, M. Kästner, DA-VEGAN: Differentiably Augmenting V AE-GAN for microstructure reconstruction from extremely small data sets, Computational Materials Science 232 (2024) 112661.doi:10.1016/j.commatsci.2023.112661. URLhttps://www.sciencedirect.com/science/article/pii/S0927025623006559

    work page doi:10.1016/j.commatsci.2023.112661 2024

  34. [34]

    Murgas, J

    B. Murgas, J. Stickel, S. Ghosh, Generative adversarial network (GAN) enabled Statistically equivalent virtual microstructures (SEVM) for modeling cold spray formed bimodal polycrystals, npj Computational Materials 10 (1) (2024) 32, publisher: Nature Publishing Group.doi:10.1038/s41524-024-01219-4. URLhttps://www.nature.com/articles/s41524-024-01219-4

    work page doi:10.1038/s41524-024-01219-4 2024

  35. [35]

    Düreth, P

    C. Düreth, P. Seibert, D. Rücker, S. Handford, M. Kästner, M. Gude, Conditional diffusion-based microstructure reconstruction, Materials Today Communications 35 (2023) 105608. doi:10.1016/j.mtcomm.2023.105608. URLhttps://www.sciencedirect.com/science/article/pii/S2352492823002982 23

    work page doi:10.1016/j.mtcomm.2023.105608 2023

  36. [36]

    K.-H. Lee, G. J. Yun, Denoising diffusion-based synthetic generation of three-dimensional (3D) anisotropic mi- crostructures from two-dimensional (2D) micrographs, Computer Methods in Applied Mechanics and Engineering 423 (2024) 116876.doi:10.1016/j.cma.2024.116876. URLhttps://www.sciencedirect.com/science/article/pii/S0045782524001324

    work page doi:10.1016/j.cma.2024.116876 2024

  37. [37]

    K.-H. Lee, G. J. Yun, Multi-plane denoising diffusion-based dimensionality expansion for 2D-to-3D reconstruction of microstructures with harmonized sampling, npj Computational Materials 10 (1) (2024) 99, publisher: Nature Publishing Group.doi:10.1038/s41524-024-01280-z. URLhttps://www.nature.com/articles/s41524-024-01280-z

    work page doi:10.1038/s41524-024-01280-z 2024

  38. [38]

    M. O. Buzzy, A. E. Robertson, S. R. Kalidindi, Statistically conditioned polycrystal generation using denoising diffusion models, Acta Materialia 267 (2024) 119746.doi:10.1016/j.actamat.2024.119746. URLhttps://www.sciencedirect.com/science/article/pii/S1359645424000995

    work page doi:10.1016/j.actamat.2024.119746 2024

  39. [39]

    Very Deep Convolutional Networks for Large-Scale Image Recognition

    K. Simonyan, A. Zisserman, Very Deep Convolutional Networks for Large-Scale Image Recognition, arXiv:1409.1556 [cs] (Apr. 2015).doi:10.48550/arXiv.1409.1556. URLhttp://arxiv.org/abs/1409.1556

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1409.1556 2015

  40. [40]

    K. He, X. Zhang, S. Ren, J. Sun, Deep Residual Learning for Image Recognition, arXiv:1512.03385 [cs] (Dec. 2015).doi:10.48550/arXiv.1512.03385. URLhttp://arxiv.org/abs/1512.03385

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1512.03385 2015

  41. [41]

    X. Li, Y . Zhang, H. Zhao, C. Burkhart, L. C. Brinson, W. Chen, A Transfer Learning Approach for Microstructure Reconstruction and Structure-property Predictions, Scientific Reports 8 (1) (2018) 13461, publisher: Nature Publishing Group.doi:10.1038/s41598-018-31571-7. URLhttps://www.nature.com/articles/s41598-018-31571-7

    work page doi:10.1038/s41598-018-31571-7 2018

  42. [42]

    Lubbers, T

    N. Lubbers, T. Lookman, K. Barros, Inferring low-dimensional microstructure representations using convolutional neural networks, Physical Review E 96 (5) (2017) 052111, publisher: American Physical Society. doi:10.1103/ PhysRevE.96.052111. URLhttps://link.aps.org/doi/10.1103/PhysRevE.96.052111

    work page doi:10.1103/physreve.96.052111 2017

  43. [43]

    Bostanabad, Reconstruction of 3D Microstructures from 2D Images via Transfer Learning, Computer-Aided Design 128 (2020) 102906.doi:10.1016/j.cad.2020.102906

    R. Bostanabad, Reconstruction of 3D Microstructures from 2D Images via Transfer Learning, Computer-Aided Design 128 (2020) 102906.doi:10.1016/j.cad.2020.102906. URLhttps://www.sciencedirect.com/science/article/pii/S0010448520300993

    work page doi:10.1016/j.cad.2020.102906 2020

  44. [44]

    Ben Britton, Tea-Sung Jun, Weimin Gan, Michael Hofmann, Fionn P.E

    P. Seibert, A. Raßloff, M. Ambati, M. Kästner, Descriptor-based reconstruction of three-dimensional microstruc- tures through gradient-based optimization, Acta Materialia 227 (2022) 117667. doi:10.1016/j.actamat. 2022.117667. URLhttps://www.sciencedirect.com/science/article/pii/S1359645422000520

    work page doi:10.1016/j.actamat 2022

  45. [45]

    Blümer, A

    V . Blümer, A. R. Safi, C. Soyarslan, B. Klusemann, T. van den Boogaard, Generative 3D reconstruction of Ti-6Al- 4V basketweave microstructures by optimization of differentiable microstructural descriptors, Acta Materialia 291 (2025) 120947.doi:10.1016/j.actamat.2025.120947. URLhttps://www.sciencedirect.com/science/article/pii/S1359645425002393

    work page doi:10.1016/j.actamat.2025.120947 2025

  46. [46]

    Kopf, C.-W

    J. Kopf, C.-W. Fu, D. Cohen-Or, O. Deussen, D. Lischinski, T.-T. Wong, Solid texture synthesis from 2D exemplars, in: ACM SIGGRAPH 2007 papers, SIGGRAPH ’07, Association for Computing Machinery, New York, NY , USA, 2007, pp. 2–es.doi:10.1145/1275808.1276380. URLhttps://dl.acm.org/doi/10.1145/1275808.1276380

    work page doi:10.1145/1275808.1276380 2007

  47. [47]

    X. Liu, V . Shapiro, Random heterogeneous materials via texture synthesis, Computational Materials Science 99 (2015) 177–189.doi:10.1016/j.commatsci.2014.12.017. URLhttps://www.sciencedirect.com/science/article/pii/S0927025614008647

    work page doi:10.1016/j.commatsci.2014.12.017 2015

  48. [48]

    A. E. Robertson, S. R. Kalidindi, Efficient Generation of Anisotropic N-Field Microstructures From 2-Point Statistics Using Multi-Output Gaussian Random Fields, SSRN Electronic Journal (2021). doi:10.2139/ssrn. 3949516. URLhttps://www.ssrn.com/abstract=3949516

    work page doi:10.2139/ssrn 2021

  49. [49]

    Senthilnathan, P

    A. Senthilnathan, P. Acar, M. De Graef, Markov Random Field based microstructure reconstruction using the principal image moments, Materials Characterization 178 (2021) 111281. doi:10.1016/j.matchar.2021. 111281. URLhttps://www.sciencedirect.com/science/article/pii/S1044580321004034

    work page doi:10.1016/j.matchar.2021 2021

  50. [50]

    Javaheri, V

    I. Javaheri, V . Sundararaghavan, Polycrystalline Microstructure Reconstruction Using Markov Random Fields and Histogram Matching, Computer-Aided Design 120 (2020) 102806.doi:10.1016/j.cad.2019.102806. URLhttps://www.sciencedirect.com/science/article/pii/S001044851930538X 24

    work page doi:10.1016/j.cad.2019.102806 2020

  51. [51]

    Seibert, M

    P. Seibert, M. Ambati, A. Raßloff, M. Kästner, Reconstructing random heterogeneous media through differen- tiable optimization, Computational Materials Science 196 (2021) 110455. doi:10.1016/j.commatsci.2021. 110455. URLhttps://www.sciencedirect.com/science/article/pii/S0927025621001804

    work page doi:10.1016/j.commatsci.2021 2021

  52. [52]

    Seibert, A

    P. Seibert, A. Raßloff, K. Kalina, M. Ambati, M. Kästner, Microstructure Characterization and Reconstruction in Python: MCRpy, Integrating Materials and Manufacturing Innovation 11 (3) (2022) 450–466. doi:10.1007/ s40192-022-00273-4. URLhttps://link.springer.com/10.1007/s40192-022-00273-4

    work page doi:10.1007/s40192-022-00273-4 2022

  53. [53]

    Seibert, A

    P. Seibert, A. Raßloff, K. A. Kalina, J. Gussone, K. Bugelnig, M. Diehl, M. Kästner, Two-stage 2D-to-3D reconstruction of realistic microstructures: Implementation and numerical validation by effective properties, Computer Methods in Applied Mechanics and Engineering 412 (2023) 116098. doi:10.1016/j.cma.2023. 116098. URLhttps://linkinghub.elsevier.com/ret...

    work page doi:10.1016/j.cma.2023 2023

  54. [54]

    Seibert, Microstructure Reconstruction from Differentiable Descriptors - A General Framework and Adapted Algorithms, Ph.D

    P. Seibert, Microstructure Reconstruction from Differentiable Descriptors - A General Framework and Adapted Algorithms, Ph.D. thesis, Technische Universität Dresden, Dresden (Apr. 2025). URLhttps://nbn-resolving.org/urn:nbn:de:bsz:14-qucosa2-963503

    2025

  55. [55]

    N. H. Paulson, M. W. Priddy, D. L. McDowell, S. R. Kalidindi, Reduced-order structure-property linkages for polycrystalline microstructures based on 2-point statistics, Acta Materialia 129 (2017) 428–438. doi: 10.1016/j.actamat.2017.03.009. URLhttps://www.sciencedirect.com/science/article/pii/S135964541730188X

    work page doi:10.1016/j.actamat.2017.03.009 2017

  56. [56]

    Benito, G

    S. Benito, G. Egels, A. Hartmaier, S. Weber, Statistical characterization of segregation-driven inhomogeneities in metallic microstructures employing fast first-order variograms, Materials Today Communications 34 (2023) 105016.doi:10.1016/j.mtcomm.2022.105016. URLhttps://www.sciencedirect.com/science/article/pii/S2352492822018578

    work page doi:10.1016/j.mtcomm.2022.105016 2023

  57. [57]

    M. D. Rintoul, S. Torquato, Reconstruction of the Structure of Dispersions, Journal of Colloid and Interface Science 186 (2) (1997) 467–476.doi:10.1006/jcis.1996.4675. URLhttps://www.sciencedirect.com/science/article/pii/S0021979796946755

    work page doi:10.1006/jcis.1996.4675 1997

  58. [58]

    Bunge, Texture Analysis in Materials Science: Mathematical Methods, Elsevier, 2013, google-Books-ID: wAQcBQAAQBAJ

    H.-J. Bunge, Texture Analysis in Materials Science: Mathematical Methods, Elsevier, 2013, google-Books-ID: wAQcBQAAQBAJ

    2013

  59. [59]

    E. G. Hemingway, O. M. O’Reilly, Perspectives on Euler angle singularities, gimbal lock, and the orthogonality of applied forces and applied moments, Multibody System Dynamics 44 (1) (2018) 31–56. doi:10.1007/ s11044-018-9620-0. URLhttps://doi.org/10.1007/s11044-018-9620-0

    work page doi:10.1007/s11044-018-9620-0 2018

  60. [60]

    Frank, Orientation Mapping: 1987 MRS Fall Meeting V on Hippel Award Lecture, MRS Bulletin 13 (3) (1988) 24–31.doi:10.1557/S0883769400066112

    C. Frank, Orientation Mapping: 1987 MRS Fall Meeting V on Hippel Award Lecture, MRS Bulletin 13 (3) (1988) 24–31.doi:10.1557/S0883769400066112. URLhttp://link.springer.com/10.1557/S0883769400066112

    work page doi:10.1557/s0883769400066112 1987

  61. [61]

    P. Neumann, Representation of Orientations of Symmetrical Objects by Rodrigues Vectors, Texture, Stress, and Microstructure 14 (1) (1991) 301034, _eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1155/TSM.14-18.53. doi:10.1155/TSM.14-18.53. URLhttps://onlinelibrary.wiley.com/doi/abs/10.1155/TSM.14-18.53

    work page doi:10.1155/tsm.14-18.53 1991

  62. [62]

    S. L. Altmann, Rotations, Quaternions, and Double Groups, Courier Corporation, 2005, google-Books-ID: Qo7NJ2kawPIC

    2005

  63. [63]

    W. R. Hamilton, On a New Species of Imaginary Quantities, Connected with the Theory of Quaternions, Proceed- ings of the Royal Irish Academy (1836-1869) 2 (1840) 424–434, publisher: Royal Irish Academy. URLhttps://www.jstor.org/stable/20520177

    arXiv

  64. [64]

    Mason, C

    J. Mason, C. Schuh, Hyperspherical harmonics for the representation of crystallographic texture, Acta Materialia 56 (20) (2008) 6141–6155.doi:10.1016/j.actamat.2008.08.031. URLhttps://linkinghub.elsevier.com/retrieve/pii/S1359645408005880

    work page doi:10.1016/j.actamat.2008.08.031 2008

  65. [65]

    Mason, C

    J. Mason, C. Schuh, Expressing Crystallographic Textures through the Orientation Distribution Function: Con- version between Generalized Spherical Harmonic and Hyperspherical Harmonic Expansions, Metallurgical and Materials Transactions A 40 (11) (2009) 2590–2602.doi:10.1007/s11661-009-9936-8. URLhttps://link.springer.com/10.1007/s11661-009-9936-8

    work page doi:10.1007/s11661-009-9936-8 2009

  66. [66]

    Mason, Quaternionic Harmonic Analysis of Texture (2012)

    J. Mason, Quaternionic Harmonic Analysis of Texture (2012). URLhttps://www.osti.gov/biblio/1231624 25

    arXiv 2012

  67. [67]

    Y . Jiao, F. H. Stillinger, S. Torquato, Modeling heterogeneous materials via two-point correlation functions: Basic principles, Physical Review E 76 (3) (2007) 031110.doi:10.1103/PhysRevE.76.031110. URLhttps://link.aps.org/doi/10.1103/PhysRevE.76.031110

    work page doi:10.1103/physreve.76.031110 2007

  68. [68]

    Y . Jiao, F. H. Stillinger, S. Torquato, Modeling heterogeneous materials via two-point correlation functions. II. Algorithmic details and applications, Physical Review E 77 (3) (2008) 031135. doi:10.1103/PhysRevE.77. 031135. URLhttps://link.aps.org/doi/10.1103/PhysRevE.77.031135

    work page doi:10.1103/physreve.77 2008

  69. [69]

    Chilès, P

    J. Chilès, P. Delfiner, Geostatistics: Modeling Spatial Uncertainty, 1st Edition, Wiley Series in Probability and Statistics, Wiley, 1999.doi:10.1002/9780470316993. URLhttps://onlinelibrary.wiley.com/doi/book/10.1002/9780470316993

    work page doi:10.1002/9780470316993 1999

  70. [70]

    Chambolle, P.-L

    A. Chambolle, P.-L. Lions, Image recovery via total variation minimization and related problems, Numerische Mathematik 76 (2) (1997) 167–188.doi:10.1007/s002110050258. URLhttps://doi.org/10.1007/s002110050258

    work page doi:10.1007/s002110050258 1997

  71. [71]

    Fletcher, Practical Methods of Optimization, 1st Edition, Wiley, 2000.doi:10.1002/9781118723203

    R. Fletcher, Practical Methods of Optimization, 1st Edition, Wiley, 2000.doi:10.1002/9781118723203. URLhttps://onlinelibrary.wiley.com/doi/book/10.1002/9781118723203

    work page doi:10.1002/9781118723203 2000

  72. [72]

    Bachmann, R

    F. Bachmann, R. Hielscher, H. Schaeben, Texture Analysis with MTEX – Free and Open Source Software Toolbox, Solid State Phenomena 160 (2010) 63–68, publisher: Trans Tech Publications Ltd. doi:10.4028/www. scientific.net/SSP.160.63. URLhttps://www.scientific.net/SSP.160.63

    work page doi:10.4028/www 2010

  73. [73]

    F. H. Stillinger, S. Torquato, Pair correlation function realizability: Lattice model implications, The Journal of Physical Chemistry B 108 (51) (2004) 19589–19594, publisher: American Chemical Society. doi:10.1021/ jp0478155. URLhttps://doi.org/10.1021/jp0478155

    work page doi:10.1021/jp0478155 2004

  74. [74]

    S. Torquato, Necessary Conditions on Realizable Two-Point Correlation Functions of Random Media, Industrial & Engineering Chemistry Research 45 (21) (2006) 6923–6928, publisher: American Chemical Society. doi: 10.1021/ie058082t. URLhttps://doi.org/10.1021/ie058082t

    work page doi:10.1021/ie058082t 2006

  75. [75]

    Barziali, J

    J. Barziali, J. M. Borwein, Two-Point Step Size Gradient Methods, IMA Journal of Numerical Analysis 8 (1) (1988) 141–148.doi:10.1093/imanum/8.1.141. URLhttps://doi.org/10.1093/imanum/8.1.141

    work page doi:10.1093/imanum/8.1.141 1988

  76. [76]

    Krause, T

    M. Krause, T. Böhlke, M. Schneider, Generating high-fidelity microstructures of polycrystalline materials with prescribed higher-order texture tensors, Computer Methods in Applied Mechanics and Engineering 452 (2026) 118690.doi:10.1016/j.cma.2025.118690. URLhttps://www.sciencedirect.com/science/article/pii/S0045782525009624

    work page doi:10.1016/j.cma.2025.118690 2026

  77. [77]

    D. P. Kingma, J. Ba, Adam: A Method for Stochastic Optimization, arXiv:1412.6980 [cs] (Jan. 2017). doi: 10.48550/arXiv.1412.6980. URLhttp://arxiv.org/abs/1412.6980

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1412.6980 2017

  78. [78]

    R. S. Dembo, T. Steihaug, Truncated-Newton algorithms for large-scale unconstrained optimization, Mathematical Programming 26 (2) (1983) 190–212.doi:10.1007/BF02592055. URLhttps://doi.org/10.1007/BF02592055

    work page doi:10.1007/bf02592055 1983

  79. [79]

    Kullback, R

    H. Robbins, S. Monro, A Stochastic Approximation Method, The Annals of Mathematical Statistics 22 (3) (1951) 400–407, publisher: Institute of Mathematical Statistics.doi:10.1214/aoms/1177729586. URL https://projecteuclid.org/journals/annals-of-mathematical-statistics/volume-22/ issue-3/A-Stochastic-Approximation-Method/10.1214/aoms/1177729586.full 26 A St...

    work page doi:10.1214/aoms/1177729586 1951