Pith. sign in

REVIEW 4 major objections 5 minor 35 references

Uncertainty-guided active learning for surrogate prediction of stream-finishing wear fields

T0 review · 4 major / 5 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read The paper shows that an uncertainty-guided surrogate can map stream-finishing wear over all 696 feasible orientations from just 92 DEM simulations, with calibrated uncertainty that predicts where the reconstruction can be trusted.

desk verdict Useful surrogate for stream finishing wear, but the active-learning benefit is not actually demonstrated without a random-sampling baseline. read the letter →

arxiv 2608.00593 v1 pith:Y3NYGPBZ submitted 2026-08-01 cond-mat.mtrl-sci cs.LGphysics.data-an

Anand Kumar , Puli Saikiran , Vineet Dawara , Koushik Viswanathan This is my paper · ORCID
classification cond-mat.mtrl-scics.LGphysics.data-an
keywords streamfinishingdiscreteelementmethodsurrogatemodeldeepensembleactivelearningepistemicuncertaintywearfieldpredictionFinnieerosion
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

Stream finishing wears a workpiece differently at each orientation, but simulating every feasible orientation with the discrete element method (DEM) is too expensive. This paper argues that a surrogate network can predict the three per-facet fields that determine erosion—normal impact velocity, tangential impact velocity, and particle flux—directly from geometric features, and that feeding these predictions through the Finnie wear formula reconstructs the entire wear-rate map. The surrogate is a deep ensemble: disagreement among its twelve members estimates where the model is uncertain, and that uncertainty selects which orientations to simulate next. Trained on 92 of 696 feasible orientations (13%), the surrogate reaches hold-out Spearman rank correlations of 0.93, 0.89, and 0.93 for the three fields, and reconstructed wear fields match DEM with rank correlation up to 0.97 for low-uncertainty orientations. If correct, orientation planning for new parts becomes affordable and comes with a trustworthy signal for where more simulation is needed.

What carries the argument

The argument is carried by combining a deep ensemble with the Finnie wear relation. Rather than regressing wear rates directly, the surrogate predicts three smooth per-triangle fields from an 11-dimensional physical feature vector: rotated centroid position, surface normal, baseline coordinates, radial distance from the container axis, and alignment of the normal with the azimuthal flow. Wear is then reconstructed analytically through the Finnie formula, which confines the sharp angular and quadratic nonlinearity to an equation. The deep ensemble's disagreement provides the epistemic uncertainty that classifies orientations into accuracy bands and selects the next DEM simulations. A farthest

What would settle it

Run DEM on 30 orientations that the final surrogate labels low-uncertainty but were never simulated, then compare predicted and DEM wear fields. The central claim fails if a substantial fraction of these orientations show Spearman rank correlations below about 0.90, or if the predicted uncertainty no longer orders the orientations by realized error.

Watch

Extended reading notes

Core claim

The central claim is that the wear-rate field of a workpiece in stream finishing is a learnable function of geometry alone, and that its smooth constituents—per-facet normal impact velocity, tangential impact velocity, and particle impact flux—can be learned from a small fraction of all orientations. The paper shows that a deep ensemble seeded with 52 farthest-point orientations and refined through five rounds of ten uncertainty-selected DEM runs predicts these fields with held-out Spearman rank correlations of 0.93, 0.89, and 0.93. The ensemble's between-member standard deviation is a truthful pre-DEM estimate of realized error, with rank correlations of -0.77 and -0.82 against R² across he

Load-bearing premise

The load-bearing premise is that each surface facet's impact velocity and flux are smooth, deterministic functions of the 11 geometric features; if unresolved many-body collision noise dominates the variation, the surrogate's sparse-sample accuracy and its uncertainty calibration will not transfer to new geometries or operating conditions.

Editorial extensions

If this is right

  • A new workpiece's full orientation-dependent wear map can be obtained from roughly a tenth of the exhaustive DEM programme, making orientation-sequence design feasible in practice.
  • The input features are computable from any triangulated mesh, so the same pipeline transfers to arbitrary complex part shapes without new input engineering.
  • Low-uncertainty orientations can be trusted without running DEM; only high-uncertainty orientations need additional simulation, and the uncertainty itself tells the user which ones.
  • Rank-based agreement remains high even where absolute wear magnitudes degrade, so decisions based on relative wear across orientations remain usable at medium uncertainty.
  • The low-uncertainty population is still growing at iteration 5, so a modest number of further batches is expected to bring nearly the whole feasible set into confident coverage.

Reading between the lines

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

  • A natural but untested extension: the uncertainty thresholds are fit to the flat-wall workpiece, so applying the same pipeline to a new geometry would require recalibration of the σ_e-to-accuracy mapping with a small validation set.
  • The 11 features are local and static; for parts with strong mutual shielding or internal passages, a facet's impact statistics may depend on the surrounding geometry, which could shrink the learnable smooth part and require more active-learning iterations.
  • Because the DEM model uses a reduced Young's modulus, the predicted wear magnitudes are relative; converting surrogate outputs to absolute wear would still require experimental calibration.
  • Choosing the next simulations by mean field-level uncertainty may be less efficient than targeting uncertainty in the reconstructed wear field or in the rank ordering of hotspots; that is a testable modification the paper does not explore.
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

4 major / 5 minor

Summary. This paper proposes a deep-ensemble surrogate to predict per-triangle normal impact velocity, tangential impact velocity, and particle impact flux for a cuboidal workpiece in stream finishing, using 11 geometric features per triangle. A farthest-point seeding scheme selects 52 initial DEM orientations, followed by five active-learning iterations that add the 10 orientations with the largest ensemble disagreement. The surrogate is evaluated on 14 held-out orientations and on uncertainty-band growth over the 696 feasible orientations. The authors report final-iteration Spearman rank correlations of 0.93, 0.89, and 0.93 for the three fields, claim the uncertainty is well calibrated, and reconstruct wear fields via the Finnie model matching DEM with Spearman correlation up to 0.97 at low uncertainty, using only 13% of the feasible orientations. The core idea is plausible, but the evidence presented does not yet establish the central claims about active-learning benefit and calibration.

Significance. The paper has several genuine strengths. The decision to predict smooth primary impact fields and to confine the nonlinear Finnie wear model to an analytical reconstruction is physically sensible and likely improves learnability. The use of an applicability-domain condition to avoid extrapolation is a thoughtful safeguard. The problem is practically relevant: orientation-dependent stream-finishing wear maps are expensive to obtain with DEM, and a reliable surrogate could substantially reduce cost. If the active-learning gains were supported by an appropriate control, the framework would be a useful template for geometry-to-wear surrogates. However, the manuscript currently stops short of demonstrating the central contribution: that uncertainty-guided acquisition outperforms simpler sampling strategies under an equal simulation budget.

major comments (4)
  1. [§2.2.6, §3.2] No baseline against random or diversity-only sampling is provided. The active-learning loop in §2.2.6 adds the 10 orientations with the largest σ_e and the abstract and §4 attribute the 13%-budget result to this uncertainty-guided strategy. But because the seed set already covers the feasible set by the d* criterion (§2.2.5), and because the training pool grows by 10 orientations per iteration, any reasonable rule that adds orientations within the covered region could plausibly produce similar gains. The reported monotone growth of the low-uncertainty band and the held-out accuracy is therefore not evidence for the value of the uncertainty signal. A control arm with the same seed set, batch size, retraining schedule, and either random or farthest-point additions is required to support the paper's central claim.
  2. [§3.1] The uncertainty calibration and its validation use the same 14 held-out orientations. Figure 5 reports Spearman correlations between σ_e and R²/ρ_s on those 14 orientations, and the σ_e thresholds for the low/medium/high bands are then fitted to the same 14-point σ_e–ρ_s relationship. This conflates calibration with validation, so the 'truthful uncertainty' claim and the band-growth figures are not independent evidence. A separate calibration split, or nested cross-validation, should be used, and the threshold values should be reported with uncertainty. With N=14, the rank correlations also have wide confidence intervals; point estimates of −0.77 and −0.82 should be accompanied by confidence intervals. In addition, §2.2.6 does not describe how the 14 validation orientations were selected; if hand-picked, this introduces another potential bias.
  3. [§3.2, Table 2.2, Figs. 6-7] No error bars or repeated realizations are reported for any main result. The deep ensemble uses random initializations and bootstrap subsets (§2.2.2), so the selected batches and final metrics are stochastic. Table 2.2 and Figures 6-7 present point estimates of R², ρ_s, and band fractions; the claims of monotone improvement and saturation at iteration 3 could reflect one realization. Re-running the loop with different ensemble seeds, or at least reporting variance over repeated training runs, is needed to support the quantitative 13%-data claim.
  4. [§3.1] The band thresholds fitted at iteration 5 are applied to earlier iterations ('the thresholds are fixed and applied unchanged to every iteration'), yet the σ_e distribution changes substantially across iterations, as shown in Table 2.2. The claim that a fixed threshold denotes the same accuracy level is reasonable only if the σ_e–ρ_s relationship is stable across iterations; this is asserted but not demonstrated. Reporting the σ_e–ρ_s relationship at each iteration, or fitting thresholds on a separate calibration set that is not also used for the Fig. 5 truthfulness test, would address this concern.
minor comments (5)
  1. [Abstract/Introduction] Typos and copyediting: 'uncertainity' in the Introduction, 'presenrlty' in the Summary, and the formatting of 'Ansys Rocky 2024R2' should be corrected.
  2. [§2.2.5] The relationship between d*, defined as the 95th percentile of nearest-neighbor distances within F, and its later use as a covering-radius stopping criterion needs clarification. Please state explicitly whether farthest-point sampling continued until the maximum distance to the nearest seed was below d*, and confirm that no feasible orientation lies beyond d* from a seed.
  3. [§3.1] The paper uses orientation-averaged R² and ρ_s, but notes that R² is unbounded below and one bad orientation can dominate an average. Reporting medians or per-orientation distributions instead of only averaged values would give a more complete picture, especially with only 14 validation orientations.
  4. [Global] No code or data availability statement is provided. Given the reproducibility value of the DEM surrogate pipeline, the authors should state whether the trained models, generated data, and scripts will be released.
  5. [§2.2.3] The 'irreducible limit' from unresolved stochastic collisions is acknowledged, but its size is never estimated. A comparison between ensemble error and the run-to-run variability of repeated DEM simulations at the same orientation would help contextualize the achievable accuracy of the surrogate.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: surrogate and wear-field predictions are trained/evaluated against independent DEM data, with no load-bearing self-citation or by-construction reduction.

full rationale

The core derivation is self-contained. DEM provides independent per-triangle ground-truth velocities and flux; the deep ensemble is trained on a subset (52→92 orientations) and evaluated on a fixed held-out set of 14 orientations excluded from training. The wear field is reconstructed by applying the analytical Finnie relation (Eq. 2.1) to the predicted primary fields, so the wear output is not used as a training target and does not reduce to the surrogate's own inputs. Epistemic uncertainty is computed from ensemble disagreement before any DEM is run for a candidate orientation, and its correlation with realized error on held-out orientations is an empirical claim that could have failed, not an identity. The only caveats are validation-design issues: the σe-to-ρs thresholds for the uncertainty bands are fitted on the same 14 held-out orientations used for the truthfulness plot, and the active-learning benefit is not compared with a random-sampling control. These are limitations in evaluation rather than circular derivations: no equation or fitted parameter is renamed as an independent prediction. References are all external; no load-bearing self-citation appears.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The central claim rests on standard ML assumptions (deep ensembles estimate useful uncertainty) and domain assumptions (DEM with reduced stiffness is a faithful ground truth; the 11-feature representation captures collision statistics). The only number 'fitted to data' in a parameter-estimation sense is the sigma_e threshold mapping, which calibrates the uncertainty bands rather than the core predictions.

free parameters (5)
  • sigma_e thresholds for uncertainty bands = (4,7)e-3 normal; (10,15)e-3 tangential
    Fit by mapping sigma_e to rho_s on the 14 held-out validation orientations at iteration 5, then fixed across iterations.
  • ensemble size M = 12
    Chosen via sensitivity analysis of uncertainty ranking stability (Fig. 3); a hyperparameter.
  • active learning iterations = 5
    Stopped for practical reasons; not derived from a convergence criterion.
  • seed set size = 52
    Determined by farthest-point sampling until covering radius below d*; d* is 95th percentile of nearest-neighbor distances.
  • validation set size = 14
    Chosen to span the feasible space; no algorithm specified.
assumptions (6)
  • domain assumption DEM with Young's modulus reduced by 10^4 preserves relative bulk flow and wear patterns (Lommen et al. [26]).
    Invoked in Section 2.1.1; if stiffness reduction alters surface impact statistics, the DEM ground truth itself is affected.
  • domain assumption Per-triangle expected impact velocities and flux are smooth, deterministic functions of the 11 geometric features.
    Stated in Sections 2.2.1 and 2.2.3; the entire interpolation-based surrogate relies on this smoothness.
  • domain assumption Epistemic uncertainty from deep ensemble disagreement estimates model uncertainty (Lakshminarayanan et al. [22]).
    Standard ML assumption adopted in Section 2.2.2; the acquisition and calibration depend on it.
  • domain assumption The 14 held-out validation orientations are representative of the 696 feasible orientations for calibrating the sigma_e-to-rho_s relation.
    Used in Section 3.1 to fix uncertainty band thresholds; a 14-point sample is a small basis for global calibration.
  • ad hoc to paper The kNN applicability domain with k=1 and the 95th percentile threshold (d*=1.21) ensures that all feasible orientations are interpolations.
    Section 2.2.5; the 95th percentile choice is conventional but arbitrary, and the interpolation guarantee is only as strong as this threshold.
  • domain assumption Finnie model applies to this workpiece material and impact regime.
    Used to reconstruct wear from predicted fields (Eq. 2.1); no validation of the erosive model against experiments is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Uncertainty-guided active learning for surrogate prediction of stream-finishing wear fields." pith.science (2026). https://pith.science/paper/Y3NYGPBZ

@misc{pith2026260800593,
  author       = {Pith},
  title        = {Pith review of: Uncertainty-guided active learning for surrogate prediction of stream-finishing wear fields},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Y3NYGPBZ}},
  note         = {Machine review of arXiv:2608.00593}
}
abstract

In stream finishing, the wear experienced by a workpiece depends strongly on its orientation within the rotating abrasive media. Determining suitable orientations to achieve uniform wear requires evaluating the wear-rate field over all feasible orientations. Although the discrete element method (DEM) accurately resolves particle interactions, simulating hundreds of feasible orientations for a new geometry is computationally expensive. We present an uncertainty-guided surrogate framework that predicts, directly from geometry, the three fields governing erosion: per-triangle normal impact velocity, tangential impact velocity, and particle impact flux. These fields are combined through the Finnie wear model to reconstruct the wear-rate distribution. The surrogate employs a deep ensemble whose disagreement estimates epistemic uncertainty, enabling an active-learning strategy that selectively performs DEM simulations for the most uncertain orientations. Trained using only $13\%$ of the $696$ feasible orientations, the surrogate achieves Spearman rank correlations of $0.93$, $0.89$, and $0.93$ for the normal impact velocity, tangential impact velocity, and particle impact flux, respectively. Moreover, the predicted uncertainty is well calibrated, reliably anticipating prediction error and the fidelity of the reconstructed wear field, which matches DEM with a Spearman rank correlation of up to $0.97$ for low-uncertainty orientations and degrades in a controlled manner as uncertainty increases.

Figures

Figures reproduced from arXiv: 2608.00593 by the authors.

Figure 1
Figure 1. Computational setup and workpiece orientations. (a) The cylindrical container holds the [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Deep-ensemble surrogate. Each of the M = 12 networks shares the same architecture but starts from a different random initialization and is trained on a different random 80% subset Dm of the seed orientations. For a given input, the M member predictions yˆ1, . . . , yˆM are combined into the ensemble mean µ∗ (the prediction used) and the between-member standard deviation σe (the epistemic uncertainty). the Adam optim… view at source ↗
Figure 3
Figure 3. Sensitivity of the orientation-level epistemic-uncertainty ranking to the ensemble size [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Overview of the active-learning surrogate framework. Farthest-point seeding selects the initial [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: Truthfulness of the predicted uncertainty on the held-out orientations of the final iteration. [PITH_FULL_IMAGE:figures/full_fig_p016_5.png]
Figure 6
Figure 6. Figure 6: Impact-velocity surrogate across the five active-learning iterations, for the normal (left column) [PITH_FULL_IMAGE:figures/full_fig_p018_6.png]
Figure 7
Figure 7. Figure 7: Particle-flux surrogate across the five active-learning iterations. (a) Orientation-averaged [PITH_FULL_IMAGE:figures/full_fig_p019_7.png]
Figure 8
Figure 8. Figure 8: Finnie-reconstructed wear fields compared with DEM for four representative orientations [PITH_FULL_IMAGE:figures/full_fig_p021_8.png]
Figure 9
Figure 9. Figure 9: Orientation-level epistemic uncertainty σe,vn (Eq. A.1) for the 644 feasible orientations not used in training at iteration 1, sorted in ascending order. The shaded band marks the highest-uncertainty orientations, which are selected for DEM simulation and added to the …
Figure 10
Figure 10. Figure 10: Per-triangle prediction and epistemic uncertainty for a representative held-out orientation. [PITH_FULL_IMAGE:figures/full_fig_p025_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 31 canonical work pages

  1. [1]

    Hashimoto.Abrasive Finishing Processes

    F. Hashimoto.Abrasive Finishing Processes. Elsevier, 2006

    work page 2006

  2. [2]

    Mass Finishing, Inc., 2007

    LK Gillespie.Mass finishing handbook. Mass Finishing, Inc., 2007

    work page 2007

  3. [3]

    Numerical simulation and experimental study of normal 25 force and particle speed in the robotic stream finishing process.Journal of Manufacturing Processes, 98:1–18, 2023

    Shengwei Ma, Keni Chih-Hua Wu, Stephen Wan, Cary Turangan, Kai Liang Tan, Wei Shin Cheng, Jun Ming Tan, and Bud Fox. Numerical simulation and experimental study of normal 25 force and particle speed in the robotic stream finishing process.Journal of Manufacturing Processes, 98:1–18, 2023

    work page 2023

  4. [4]

    P. A. Cundall and O. D. L. Strack. A discrete numerical model for granular assemblies. G´ eotechnique, 29(1):47–65, 1979

    work page 1979

  5. [5]

    I. Finnie. Erosion of surfaces by solid particles.Wear, 3(2):87–103, 1960

    work page 1960

  6. [6]

    H. P. Zhu, Z. Y. Zhou, R. Y. Yang, and A. B. Yu.Discrete Particle Simulation of Particulate Systems: Theoretical Developments, volume 63. Chemical Engineering Science, 2008

    work page 2008

  7. [7]

    Agent-based discrete element modeling of microbial-induced carbonate precipitation.Advanced Theory and Simulations, 9(3):e02233, 2026

    Liming Zhao, Brooke Filanoski, Ruohong Chen, David Erickson, and Jingjie Yeo. Agent-based discrete element modeling of microbial-induced carbonate precipitation.Advanced Theory and Simulations, 9(3):e02233, 2026

    work page 2026

  8. [8]

    P. W. Cleary. Large scale industrial dem modelling.Engineering Computations, 21(2/3/4):169– 204, 2004

    work page 2004

Show all 35 references
  1. [9]

    Calibrated and validated wear prediction for bulk material handling equipment using dem simulations

    Thomas Roessler and Andre Katterfeld. Calibrated and validated wear prediction for bulk material handling equipment using dem simulations. InICBMH2023: 14th International Conference on Bulk Materials Storage, Handling and Transportation, pages 284–295. The Institution of Engin...

    2023

  2. [10]

    Simulation of solid particle erosion wear using discrete element method: Comparison of experimental and analysis results.Particuology, 2025

    Mehmet Esat Aydin, Veysel Firat, and Mehmet Bagci. Simulation of solid particle erosion wear using discrete element method: Comparison of experimental and analysis results.Particuology, 2025

    2025

  3. [11]

    Optimization of the stream finishing process for mechanical surface treatment by numerical and experimental process analysis.CIRP Annals, 68(1):373–376, 2019

    Frederik Zanger, Andreas Kacaras, Patrick Neuenfeldt, and Volker Schulze. Optimization of the stream finishing process for mechanical surface treatment by numerical and experimental process analysis.CIRP Annals, 68(1):373–376, 2019

    2019

  4. [12]

    Grinding surface roughness measurement combined with simulation data and transfer learning.Advanced Theory and Simulations, 7(5):2301100, 2024

    Huaian Yi, Ziqiang Feng, and Aihua Shu. Grinding surface roughness measurement combined with simulation data and transfer learning.Advanced Theory and Simulations, 7(5):2301100, 2024. 26

    2024

  5. [13]

    Efficient global optimization of expensive black-box functions.Journal of Global Optimization, 13(4):455–492, 1998

    Donald R Jones, Matthias Schonlau, and William J Welch. Efficient global optimization of expensive black-box functions.Journal of Global Optimization, 13(4):455–492, 1998

    1998

  6. [14]

    Liu, and X

    Kaifeng Yang, S. Liu, and X. Chen. Bayesian optimization with gaussian process surrogates for high-dimensional design problems.Applied Mathematical Modelling, 104:215–231, 2022

    2022

  7. [15]

    Kriging-based design optimization of an injection molding process for a light guide plate.Journal of Mechanical Science and Technology, 24(1):97–100, 2010

    Soo-Hoon Chae, Hyo-Jin Kim, and Myung-Woo Cho. Kriging-based design optimization of an injection molding process for a light guide plate.Journal of Mechanical Science and Technology, 24(1):97–100, 2010

    2010

  8. [16]

    Efficient global optimization applied to aerodynamic design of flyback booster.Journal of Spacecraft and Rockets, 44(5):1022–1030, 2007

    Masahiro Kanazaki, Shinkyu Jeong, and Shigeru Obayashi. Efficient global optimization applied to aerodynamic design of flyback booster.Journal of Spacecraft and Rockets, 44(5):1022–1030, 2007

    2007

  9. [17]

    Jasper Snoek, Hugo Larochelle, and Ryan P. Adams. Practical bayesian optimization of machine learning algorithms. InAdvances in Neural Information Processing Systems, volume 25, pages 2951–2959, 2012

    2012

  10. [18]

    Bayesian methods in political science

    Jong Hee Park and Sooahn Shin. Bayesian methods in political science. In Luigi Curini and Robert Franzese, editors,The SAGE Handbook of Research Methods in Political Science & International Relations, pages 895–909. SAGE Publications, 2020

    2020

  11. [19]

    Kriging-surrogate-based optimiza- tion considering expected hypervolume improvement in non-constrained many-objective test problems

    Koji Shimoyama, Shinkyu Jeong, and Shigeru Obayashi. Kriging-surrogate-based optimiza- tion considering expected hypervolume improvement in non-constrained many-objective test problems. In2013 IEEE Congress on Evolutionary Computation, pages 2262–2269. IEEE, 2013

    2013

  12. [20]

    Weight uncer- tainty in neural networks

    Charles Blundell, Julien Cornebise, Koray Kavukcuoglu, and Daan Wierstra. Weight uncer- tainty in neural networks. InProceedings of the 32nd International Conference on Machine Learning (ICML), volume 37, pages 1613–1622, 2015

    2015

  13. [21]

    Dropout as a bayesian approximation: Representing model uncertainty in deep learning

    Yarin Gal and Zoubin Ghahramani. Dropout as a bayesian approximation: Representing model uncertainty in deep learning. InProceedings of the 33rd International Conference on Machine Learning (ICML), volume 48, pages 1050–1059, 2016. 27

    2016

  14. [22]

    Simple and scalable predictive uncertainty estimation using deep ensembles

    Balaji Lakshminarayanan, Alexander Pritzel, and Charles Blundell. Simple and scalable predictive uncertainty estimation using deep ensembles. InAdvances in Neural Information Processing Systems, volume 30, 2017

    2017

  15. [23]

    Sculley, Sebastian Nowozin, Joshua V

    Yaniv Ovadia, Emily Fertig, Jie Ren, Zachary Nado, D. Sculley, Sebastian Nowozin, Joshua V. Dillon, Balaji Lakshminarayanan, and Jasper Snoek. Can you trust your model’s uncertainty? evaluating predictive uncertainty under dataset shift. InAdvances in Neural Information Proces...

    2019

  16. [24]

    Deep ensembles: A loss landscape perspective

    Stanislav Fort, Huiyi Hu, and Balaji Lakshminarayanan. Deep ensembles: A loss landscape perspective. arxiv 2019.arXiv preprint arXiv:1912.02757, 2019

    2019 arXiv

  17. [25]

    Mindlin and Herbert Deresiewicz

    Raymond D. Mindlin and Herbert Deresiewicz. Elastic spheres in contact under varying oblique forces.Journal of Applied Mechanics, 20(3):327–344, 1953

    1953

  18. [26]

    Dem speedup: Stiffness effects on behavior of bulk material.Particuology, 12:107–112, 2014

    Stef Lommen, Dingena Schott, and Gabriel Lodewijks. Dem speedup: Stiffness effects on behavior of bulk material.Particuology, 12:107–112, 2014

    2014

  19. [27]

    Kingma and Jimmy Ba

    Diederik P. Kingma and Jimmy Ba. Adam: A method for stochastic optimization. In3rd International Conference on Learning Representations (ICLR), 2015

    2015

  20. [28]

    Active learning literature survey

    Burr Settles. Active learning literature survey. Technical Report 1648, University of Wisconsin– Madison, 2009

    2009

  21. [29]

    Physics-guided architecture (pga) of neural networks for quantifying uncertainty in lake temperature modeling

    Arka Daw, R Quinn Thomas, Cayelan C Carey, Jordan S Read, Alison P Appling, and Anuj Karpatne. Physics-guided architecture (pga) of neural networks for quantifying uncertainty in lake temperature modeling. InProceedings of the 2020 siam international conference on data mining,...

    2020

  22. [30]

    Defining a novel k-nearest neighbours approach to assess the applicability domain of a qsar model for reliable predictions.Journal of cheminformatics, 5(1):27, 2013

    Faizan Sahigara, Davide Ballabio, Roberto Todeschini, and Viviana Consonni. Defining a novel k-nearest neighbours approach to assess the applicability domain of a qsar model for reliable predictions.Journal of cheminformatics, 5(1):27, 2013. 28

    2013

  23. [31]

    Out-of-distribution detection with deep nearest neighbors

    Yiyou Sun, Yifei Ming, Xiaojin Zhu, and Yixuan Li. Out-of-distribution detection with deep nearest neighbors. InInternational conference on machine learning, pages 20827–20840. PMLR, 2022

    2022

  24. [32]

    Gonzalez

    Teofilo F. Gonzalez. Clustering to minimize the maximum intercluster distance.Theoretical Computer Science, 38:293–306, 1985

    1985

  25. [33]

    Active learning for convolutional neural networks: A core-set approach.arXiv preprint arXiv:1708.00489, 2017

    Ozan Sener and Silvio Savarese. Active learning for convolutional neural networks: A core-set approach.arXiv preprint arXiv:1708.00489, 2017

    2017 arXiv

  26. [34]

    Spearman

    C. Spearman. The proof and measurement of association between two things.American Journal of Psychology, 15(1):72–101, 1904

    1904

  27. [35]

    Fatigue life prediction of glare composites using regression tree ensemble-based machine learning model.Advanced Theory and Simulations, 3(6):2000048, 2020

    Wei Sai, Gin Boay Chai, and Narasimalu Srikanth. Fatigue life prediction of glare composites using regression tree ensemble-based machine learning model.Advanced Theory and Simulations, 3(6):2000048, 2020. 29

    2020

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.