Pith. sign in

REVIEW 3 major objections 5 minor 32 references

Robust Graph-Based Semi-Supervised Learning via $p$-Conductances

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read A family of p-conductance energy programs, solved with heat-kernel diffused labels, is claimed to achieve state-of-the-art accuracy in graph-based semi-supervised learning under scarce, corrupted, and partial labels.

desk verdict The gauge-duality core and p-conductance family are a real contribution, but the experimental section's self-contradictory epsilon definitions make the headline accuracy claims unreadable as written. read the letter →

arxiv 2502.08873 v1 pith:X7SANQHL submitted 2025-02-13 cs.LG cs.DMmath.OC

Sawyer Jack Robertson , Chester Holtz , Zhengchao Wan , Gal Mishne , Alexander Cloninger This is my paper · ORCID
classification cs.LGcs.DMmath.OC MSC 68Q8790C3505C9090C53
keywords semi-supervisedlearningnodeclassificationp-Laplaciangrapheffectiveresistanceoptimaltransportlabelcorruptionheatkernelmeasuremincut
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper proposes p-conductance learning, a graph-based semi-supervised method that replaces hard label constraints with an affine constraint on two label probability measures. The central object is the energy program $C_p(\mu,\nu)=\min_{\phi:\phi^T(\mu-\nu)=1}(\sum_{ij} w_{ij}|\phi_i-\phi_j|^p)^{1/p}$, which interpolates between sparse-cut behavior at $p=1$, effective-resistance behavior at $p=2$, and Wasserstein/transport behavior at $p=\infty$. The authors argue that diffusing the labels with the heat kernel before solving makes the resulting sign predictions robust to label corruption, and they support this with a stability bound for $p=2$ plus experiments on citation and image datasets. Sympathetically read, the paper's claim is that one tunable energy family, solved by a fast semismooth Newton method, achieves state-of-the-art accuracy in exactly the regimes where labels are scarce, noisy, or only partially known.

What carries the argument

The load-bearing object is the measure p-conductance $C_p(\mu,\nu)$, a variational program in which the unknown potential $\phi$ is coupled to the two label measures through the affine constraint $\phi^T(\mu-\nu)=1$ instead of hard node-wise label equalities. The paper shows via gauge duality that $C_p$ is the reciprocal of a p-Beckmann optimal-transport metric, which yields the effective-resistance identity at $p=2$ and the Wasserstein identity at $p=\infty$. Robustness is carried by the heat-kernel diffusion operator $e^{-tL}$ applied to the label measures, combined with a spectral bound on $L^+$; Theorem 2.7 bounds the $\ell^2$ change in the potential by $t\|\mu-\nu\|_2 + \lambda^{-1} e^{-t\lambda} \|\eta\|_2$. Computationally, the paper introduces a semismooth Newton conjugate-gradient augmented-Lagrangian method for all $p\in[1,\infty]$, whose convergence rests on strong semismoothness of the proximal map of the weighted p-norm.

What would settle it

Take a graph and label noise $\eta$ satisfying the Theorem 2.7 condition $\|\eta\|_2>\|\mu-\nu\|_2$, and compute the fraction of nodes where sign($\phi$) changes between the clean and diffused corrupted potentials. If a small $\ell^2$ perturbation flips the sign at a large fraction of high-confidence nodes, or if increasing $t$ decreases accuracy rather than increasing it, the robustness mechanism described in the paper is not what drives the empirical gains.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that the measure p-conductance program $C_p$ — an affine-relaxed, p-Laplacian-style objective — is the right interpolation between Poisson learning and p-Laplace learning, and that heat-kernel diffused labels make it robust. For $p=2$ the minimizer is a rescaling of the Poisson potential, $\phi^*=L^+(\mu-\nu)/((\mu-\nu)^T L^+ (\mu-\nu))$, so the method literally generalizes Poisson learning. The paper proves that $C_1$ is a generalized min-cut/max-flow problem, that $C_2^2$ is the reciprocal of measure effective resistance, and that $C_\infty$ is the reciprocal of a 1-Wasserstein distance. It also proves that for $p=2$, applying $e^{-tL}$ to the labels improves a worst-case $\ell^2$ bound on the potential when label noise is large relative to the clean signal. Empirically, the method reports higher accuracy than Poisson learning and PoissonMBO in low label-rate, corrupted-label, and partial-label settings on Cora, Citeseer, Pubmed, MNIST, FashionMNIST, CIFAR-10, and CIFAR-100.

Load-bearing premise

The paper assumes that a small $\ell^2$ perturbation of the continuous potential (bounded in Theorem 2.7) translates into unchanged sign or argmax predictions; the link between that bound and classification accuracy is not proved, and the diffusion time $t$ is chosen empirically.

Editorial extensions

If this is right

  • At $p=2$ with no cardinality prior ($\epsilon=0$), p-conductance learning reports the best or tied-best accuracy on Cora and Pubmed across label rates from 1 to 100 labels per class, including a 4.6% gain over PoissonMBO at 1 label per class on Cora.
  • On CIFAR-10 with 40% flipped labels, the method reports a 3.8% accuracy improvement over Poisson learning, and the experiments show the gain depends on heat-kernel diffusion time.
  • In the CIFAR-100 partially-labeled setting with a candidate set of size 4, p-conductance-MBO outperforms PoissonMBO by 4.1%.
  • The same energy family covers $p=1,2,\infty$, giving a unified view: sparse cuts at $p=1$, effective resistance at $p=2$, and transport geometry at $p=\infty$.

Reading between the lines

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

  • The stability bound in Theorem 2.7 concerns the $\ell^2$ distance of potentials, not the sign or argmax of predictions; a natural extension is to prove that, under a margin or separation condition on the clean potential, the same diffusion step also stabilizes the label error rate, which would close the gap between the theorem and the empirical robustness claims.
  • Because $C_\infty$ is the reciprocal of a shortest-path Wasserstein distance, the p-conductance family can be read as a one-parameter interpolation between cut geometry and transport geometry; this suggests testable connections between label robustness and the coarse geometry of the graph, e.g., graphs with larger spectral gap should show smaller accuracy drops under noise.
  • The diffusion time $t$ is treated as a free parameter and chosen empirically in Figure 3; one could derive a data-dependent rule for $t$ from the bound in Theorem 2.7, e.g., $t\approx(1/\lambda)(\|\eta\|/\|\mu-\nu\| - 1)$, and test whether the automatically chosen $t$ matches the empirical optimum.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proposes p-conductance learning, a graph-based SSL method that solves a measure-constrained p-Dirichlet energy minimization problem with an affine label constraint. It derives connections to mincut, randomized cuts, normalized cuts, effective resistance, and Wasserstein distance via gauge duality, proves an ℓ2 stability bound for diffused labels at p=2, develops a semismooth Newton augmented-Lagrangian algorithm, and reports experiments on citation and image datasets claiming state-of-the-art accuracy in low-label-rate, corrupted-label, and partial-label regimes.

Significance. If the claims hold, the paper gives a clean variational framework that unifies p-Laplace and Poisson learning with classical graph-cut and optimal-transport objects; the explicit p=2 minimizer in Eq. (8) and the gauge-duality relationships in Theorem 2.4 are attractive and appear correct. The SSNAL algorithm is a standard and well-cited extension of existing techniques. However, the empirical headline is currently not verifiable: the meaning of the cardinality parameter ε is stated inconsistently between Section 4 and Appendix D.1, the experimental configuration (p, t, ε) is under-specified, and no code or data is provided. In addition, Theorem 2.7 bounds the ℓ2 perturbation of the potential but does not imply sign or argmax preservation, so the theoretical support for the robustness claim is weaker than the abstract suggests.

major comments (3)
  1. [Section 4 and Appendix D.1] Section 4 states: 'we denote by p-conductance (ε = 0) to be our method without any cardinality prior and p-conductance (ε = n) to be our method with exact knowledge of class cardinalities.' Appendix D.1, Eq. (28), relaxes the cardinality constraint to m−ε ≤ Pᵀ1 ≤ m+ε, and Remark D.1 states that ε=0 is the LP over the transportation polytope with exact cardinalities, while ε=n corresponds to the unconstrained argmax heuristic. These two statements are exact opposites. Because the tables (e.g., Table 1, Cora with 1 label per class, 63.1 for ε=0 versus 58.9 for ε=n) use these labels without code or data, a reader cannot determine which configuration produced the reported state-of-the-art numbers. The convention must be fixed and the experimental tables re-labeled or re-run accordingly.
  2. [Section 2.2, Theorem 2.7] Theorem 2.7 bounds ∥ψ − ψ~t∥₂ in terms of the perturbation η and the diffusion time t, but the method's predictions are sgn(φ*) or argmax over columns of φ*. An ℓ2 bound on the potential does not imply that signs or argmax orderings are preserved; no margin or separation assumption is stated. Moreover, Remark 2.8 shows that the improvement condition requires ∥η∥₂/∥µ−ν∥₂ > 1, which in the binary example means more than one quarter of the labels are corrupted, so the theorem does not cover the small-corruption regime shown in the experiments. The diffusion time t is also chosen empirically in Figure 3 rather than prescribed by the theorem. As a result, the theoretical result only motivates, rather than establishes, the abstract's robustness claim, and the gap between the ℓ2 statement and classification accuracy should be addressed explicitly.
  3. [Section 4, Tables 1–6] The experimental configuration is under-specified. The values of p, the diffusion time t, and the cardinality slack ε are not reported for each dataset and label rate; only the graph construction parameter k=10 is fixed for image datasets. Furthermore, for p=2, Eq. (8) shows that the optimal potential is a positive scalar multiple of the Poisson learning potential L†(µ−ν). Consequently, the gains reported for p=2 over the Poisson baseline could come entirely from the heat-kernel diffusion and the cardinality-aware rounding in Eq. (28), rather than from the new variational objective. An ablation separating these components (for example, Poisson learning with diffused labels, and p=2 with and without cardinality rounding) is needed to support the attribution in the abstract, and the absence of code or data makes these variants impossible to reconstruct.
minor comments (5)
  1. [Theorem 3.7] The convergence-rate statement is missing a closing parenthesis; it should read ‖φ^{k+1} − φ̄‖ = O(‖φ^k − φ̄‖^{1+τ}).
  2. [Proposition 3.5] The proposition states 'integer values of p ∈ [0, ∞)', but the problem (Cp) is defined for p ≥ 1; p=0 is not used anywhere.
  3. [Table 6] The CIFAR-100 PLL table lists 'p-conductance (p = 2, ε = n)' twice with different accuracy values; one of these rows should presumably be labeled with the other ε value used in the paper.
  4. [Appendix D.1, Eq. (29)] Equation (29) writes 'φ∗_i = arg max ...', but the left-hand side should be the predicted class label, not the optimal potential.
  5. [Section A.1, Table 2] The text says 'p-conductance learning outperforms all other methods' in the image-dataset section, but on MNIST p-conductance-MBO ties PoissonMBO exactly at every reported label count (97.5, 97.5, 97.5, 97.6, 98.1); the claim should be softened to 'matches or outperforms'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the derivation chain runs from the Cp program through gauge duality and standard Laplacian identities, the p=2 reduction to Poisson learning is explicitly acknowledged, and no fitted parameter is renamed as a prediction.

full rationale

Walking the derivation chain, I find no step in which an output quantity is defined in terms of itself, a fitted parameter is renamed as a prediction, or a load-bearing premise rests solely on a same-author citation. The main theoretical results are derived in-text: Theorem 2.1 is LP duality; Theorem 2.2 and Remark 2.3 are elementary reformulations; Theorem 2.4 follows from gauge duality (Friedlander et al. 2014) with the norm-polar computation in Lemma B.1; Corollary 2.6 then uses the standard identity B_{w^{-1},2}^2 = (mu-nu)^T L^dagger (mu-nu). Equation (8) does show that for p=2 the minimizer is a positive scalar multiple of L^dagger (mu-nu), so sign predictions coincide with Poisson learning, but the paper states this connection explicitly (it says the method generalizes Poisson and p-Laplace learning) and does not present that equivalence as an independent prediction; this is a novelty overlap, not circularity. Theorem 2.7 is an independent norm bound (lambda^{-1} e^{-t lambda} ||eta|| + t ||mu-nu||), although it does not by itself imply sign or argmax stability, which is a correctness gap rather than a circularity. The self-citation to Robertson et al. (2024) is contextual, and Theorem 2.4 is re-proved here rather than imported as a black box. The contradictory uses of epsilon in Section 4 versus Remark D.1 are an internal consistency problem for interpreting the experimental tables, not a case of a result reducing to its input. No fitted parameter is relabeled as a prediction; the diffusion time t is chosen empirically and is not presented as a theoretically forced value. Overall, the central derivations are self-contained and do not assume the target results.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central method rests on standard convex and norm duality for the theoretical links, a connected-graph spectral assumption for the stability bounds, and the ad hoc modeling choices of diffused label measures and thresholding or cardinality assignment. No new physical entities are introduced; the main free parameters are p, diffusion time t, the cardinality slack epsilon, and graph construction parameters.

free parameters (4)
  • p = 1, 2, or infinity; headline results use p=2
    Exponent in the p-conductance objective controls cut sparsity; experiments report all three values but the state-of-the-art claims are carried by p=2. No theory fixes p for a given dataset.
  • diffusion time t = positive, chosen per dataset and swept in Figure 3
    Labels are preprocessed as e^{-tL} Y; the robustness theorem only gives a range where worst-case potential error improves, not a rule for selecting t.
  • epsilon (cardinality slack) = 0 or n
    Controls whether the assignment LP enforces class-size estimates; the paper's Remark 4.1 and Appendix D.1 disagree about which value means exact cardinality, so the experimental variant labels are ambiguous.
  • k-nearest neighbors k = 10
    Graph construction parameter for image datasets; fixed in all experiments, affects graph structure and all downstream results.
assumptions (4)
  • standard math Gauge duality and strong duality for linearly constrained norm minimization.
    Used in Theorem 2.4 and Appendix B.1 to relate Cp to Beckmann metrics; relies on Friedlander et al. (2014).
  • domain assumption The graph is connected and the Laplacian has a positive spectral gap.
    Assumed in Theorems 2.1, 2.4, 2.7 and Remark 3.2 for strong duality, pseudoinverse bounds, and positive-definiteness; experiments restrict to largest connected components.
  • ad hoc to paper Label information can be encoded as probability measures mu and nu, and predictions are obtained by thresholding or argmax of the optimal potential.
    This modeling choice defines the method; it is not derived from data and its validity is supported only empirically.
  • ad hoc to paper One global heat-kernel diffusion time t improves robustness.
    Used in Section 4 and Figure 3; the theorem gives only a conditional worst-case guarantee, not a prescription for t.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Robust Graph-Based Semi-Supervised Learning via $p$-Conductances." pith.science (2026). https://pith.science/paper/X7SANQHL

@misc{pith2026250208873,
  author       = {Pith},
  title        = {Pith review of: Robust Graph-Based Semi-Supervised Learning via $p$-Conductances},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/X7SANQHL}},
  note         = {Machine review of arXiv:2502.08873}
}
abstract

We study the problem of semi-supervised learning on graphs in the regime where data labels are scarce or possibly corrupted. We propose an approach called $p$-conductance learning that generalizes the $p$-Laplace and Poisson learning methods by introducing an objective reminiscent of $p$-Laplacian regularization and an affine relaxation of the label constraints. This leads to a family of probability measure mincut programs that balance sparse edge removal with accurate distribution separation. Our theoretical analysis connects these programs to well-known variational and probabilistic problems on graphs (including randomized cuts, effective resistance, and Wasserstein distance) and provides motivation for robustness when labels are diffused via the heat kernel. Computationally, we develop a semismooth Newton-conjugate gradient algorithm and extend it to incorporate class-size estimates when converting the continuous solutions into label assignments. Empirical results on computer vision and citation datasets demonstrate that our approach achieves state-of-the-art accuracy in low label-rate, corrupted-label, and partial-label regimes.

Figures

Figures reproduced from arXiv: 2502.08873 by the authors.

Figure 1
Figure 1. (Left) One-vs-all measure mincut on the Sklearn Digits dataset Dua & Graff (2017). The initially labeled nodes are shown with images overlaid. µ is given by the five images of the digit six, and ν from all of the other classes. Solving the program (Cp) for p = 1, we obtain a sparse solution ϕ such that each {i, j} ∈ E satisfies |ϕi − ϕj | ≈ 0 (light gray) or |ϕi − ϕj | ≈ 0.11 (red). (Right) Removing red edges isolat… view at source ↗
Figure 2
Figure 2. Visualizing solutions ϕ to Cp for p ∈ [1, ∞]. Measures µ, ν appear in blue and red, with opacity proportional to value. Edge width is proportional to |ϕi − ϕj |. “X” symbols mark edges crossing {ϕ ≤ T} and {ϕ > T}, where T is the mean of ϕ. Here, Be ∈ Rn×2N is the oriented incidence matrix according to E′ . Finally, strong duality holds. When only one node is labeled in each class, mincut(µ, ν) recovers the usual no… view at source ↗
Figure 3
Figure 3. Accuracy on CIFAR-10 for various diffusion times and robustness. Shades denote noise [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Convergence of KKT residuals on lattice data for [PITH_FULL_IMAGE:figures/full_fig_p020_4.png]
Figure 5
Figure 5. Figure 5: MNIST experiments 20 [PITH_FULL_IMAGE:figures/full_fig_p020_5.png]
Figure 6
Figure 6. Figure 6: FashionMNIST experiments. The labels are 0: T-shirt/top 1: Trouser 2: Pullover 3: [PITH_FULL_IMAGE:figures/full_fig_p021_6.png]
Figure 7
Figure 7. Figure 7: An illustration of the one-vs-all minimum measure cut on a graph obtained from the [PITH_FULL_IMAGE:figures/full_fig_p021_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 25 canonical work pages

  1. [1]

    Fast, provably convergent irls algorithm for p-norm linear regression

    Adil, D., Peng, R., and Sachdeva, S. Fast, provably convergent irls algorithm for p-norm linear regression. In Wallach, H., Larochelle, H., Beygelzimer, A., d Alch\' e -Buc, F., Fox, E., and Garnett, R. (eds.), Advances in Neural Information Processing Systems, volume 32. Curran Associates, Inc., 2019. URL https://proceedings.neurips.cc/paper_files/paper/...

    work page 2019

  2. [2]

    Alaoui, A. E. K., Cheng, X., Ramdas, A., Wainwright, M. J., and Jordan, M. I. Asymptotic behavior of p-based laplacian regularization in semi-supervised learning. ArXiv, abs/1603.00564, 2016. URL https://api.semanticscholar.org/CorpusID:1089782

    work page Pith review arXiv 2016

  3. [3]

    A continuous model of transportation

    Beckmann, M. A continuous model of transportation. Econometrica: Journal of the Econometric Society, 1952

    work page 1952

  4. [4]

    Consistency of lipschitz learning with infinite unlabeled data and finite labeled data

    Calder, J. Consistency of lipschitz learning with infinite unlabeled data and finite labeled data. SIAM Journal on Mathematics of Data Science, 1 0 (4): 0 780--812, 2019. doi:10.1137/18M1199241. URL https://doi.org/10.1137/18M1199241

    doi:10.1137/18m1199241 2019

  5. [5]

    Graphlearning python package, January 2022

    Calder, J. Graphlearning python package, January 2022. URL https://doi.org/10.5281/zenodo.5850940

    doi:10.5281/zenodo.5850940 2022

  6. [6]

    and Ettehad, M

    Calder, J. and Ettehad, M. Hamilton-jacobi equations on graphs with applications to semi-supervised learning and data depth. J. Mach. Learn. Res., 23 0 (1), January 2022. ISSN 1532-4435

    work page 2022

  7. [7]

    Poisson learning: Graph based semi-supervised learning at very low label rates

    Calder, J., Cook, B., Thorpe, M., and Slep c ev, D. Poisson learning: Graph based semi-supervised learning at very low label rates. In Proceedings of the 37th International Conference on Machine Learning, ICML'20. JMLR.org, 2020

    work page 2020

  8. [8]

    The MNIST database of handwritten digit images for machine learning research

    Deng, L. The MNIST database of handwritten digit images for machine learning research. IEEE Signal Processing Magazine, 29 0 (6): 0 141--142, 2012

    work page 2012

Show all 32 references
  1. [9]

    and Graff, C

    Dua, D. and Graff, C. UCI machine learning repository, 2017. URL http://archive.ics.uci.edu/ml

    2017

  2. [10]

    Algorithms for semisupervised learning on graphs

    Flores, M. Algorithms for semisupervised learning on graphs. 2018

    2018

  3. [11]

    Analysis and algorithms for p-based semi-supervised learning on graphs

    Flores, M., Calder, J., and Lerman, G. Analysis and algorithms for p-based semi-supervised learning on graphs. Applied and Computational Harmonic Analysis, 60: 0 77--122, 2022. ISSN 1063-5203. doi:https://doi.org/10.1016/j.acha.2022.01.004. URL https://www.sciencedirect.com/sc...

    2022 doi

  4. [12]

    Freund, R. M. Dual gauge programs, with applications to quadratic programming and the minimum-norm problem. Mathematical Programming, 38, 1987

    1987

  5. [13]

    P., Macedo, I., and Pong, T

    Friedlander, M. P., Macedo, I., and Pong, T. K. Gauge optimization and duality. SIAM Journal on Optimization, 24 0 (4), 2014

    2014

  6. [14]

    Continuous partitioning for graph-based semi-supervised learning

    Holtz, C., Chen, P., Wan, Z., Cheng, C.-K., and Mishne, G. Continuous partitioning for graph-based semi-supervised learning. In The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024. URL https://openreview.net/forum?id=hCOuip5Ona

    2024

  7. [15]

    Auction dynamics: A volume constrained MBO scheme

    Jacobs, M., Merkurjev, E., and Esedoḡlu, S. Auction dynamics: A volume constrained MBO scheme. Journal of Computational Physics, 354: 0 288--310, 2018. ISSN 0021-9991. doi:https://doi.org/10.1016/j.jcp.2017.10.036. URL https://www.sciencedirect.com/science/article/pii/S0021999...

    2018 doi

  8. [16]

    Jung, A., III, A. O. H., Mara, A., and Jahromi, S. Semi-supervised learning via sparse label propagation, 2016

    2016

  9. [17]

    Kingma, D. P. and Welling, M. Auto-Encoding Variational Bayes . In 2nd International Conference on Learning Representations, ICLR 2014, Banff, AB, Canada, April 14-16, 2014, Conference Track Proceedings , 2014

    2014

  10. [18]

    and Hinton, G

    Krizhevsky, A. and Hinton, G. Learning multiple layers of features from tiny images. 0 (0), 2009

    2009

  11. [19]

    Kyng, R., Rao, A., Sachdeva, S., and Spielman, D. A. Algorithms for lipschitz learning on graphs. In Grünwald, P., Hazan, E., and Kale, S. (eds.), Proceedings of The 28th Conference on Learning Theory, volume 40 of Proceedings of Machine Learning Research, pp.\ 1190--1223, Par...

    2015

  12. [20]

    Loera, J. A. D. and Kim, E. D. Combinatorics and geometry of transportation polytopes: An update, 2013. URL https://arxiv.org/abs/1307.0124

    2013 arXiv

  13. [21]

    Semi-supervised learning with the graph laplacian: the limit of infinite unlabelled data

    Nadler, B., Srebro, N., and Zhou, X. Semi-supervised learning with the graph laplacian: the limit of infinite unlabelled data. In Proceedings of the 23rd International Conference on Neural Information Processing Systems, NIPS'09, pp.\ 1330–1338, Red Hook, NY, USA, 2009. Curran...

    2009

  14. [22]

    Computational optimal transport: With applications to data science

    Peyr \'e , G., Cuturi, M., et al. Computational optimal transport: With applications to data science. Foundations and Trends in Machine Learning , 11 0 (5-6), 2019

    2019

  15. [23]

    All you need is resistance: On the equivalence of effective resistance and certain optimal transport problems on graphs

    Robertson, S., Wan, Z., and Cloninger, A. All you need is resistance: On the equivalence of effective resistance and certain optimal transport problems on graphs. arXiv preprint arXiv:2404.15261, 2024

    2024 arXiv

  16. [24]

    Rockafellar, R. T. Augmented lagrangians and applications of the proximal point algorithm in convex programming. Mathematics of Operations Research, 1 0 (2): 0 97--116, 1976 a . ISSN 0364765X, 15265471. URL http://www.jstor.org/stable/3689277

    1976

  17. [25]

    Rockafellar, R. T. Monotone operators and the proximal point algorithm. SIAM Journal on Control and Optimization, 14 0 (5): 0 877--898, 1976 b . doi:10.1137/0314056. URL https://doi.org/10.1137/0314056

    1976 doi

  18. [26]

    and Herbster, M

    Saito, S. and Herbster, M. Multi-class graph clustering via approximated effective p -resistance. In International Conference on Machine Learning, pp.\ 29697--29733. PMLR, 2023

    2023

  19. [27]

    Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms

    Xiao, H., Rasul, K., and Vollgraf, R. Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms. ArXiv, abs/1708.07747, 2017

    2017 arXiv

  20. [28]

    W., and Salakhutdinov, R

    Yang, Z., Cohen, W. W., and Salakhutdinov, R. Revisiting semi-supervised learning with graph embeddings. In Proceedings of the 33rd International Conference on International Conference on Machine Learning - Volume 48, ICML'16, pp.\ 40–48. JMLR.org, 2016

    2016

  21. [29]

    A newton-cg augmented lagrangian method for semidefinite programming

    Zhao, X.-Y., Sun, D., and Toh, K.-C. A newton-cg augmented lagrangian method for semidefinite programming. SIAM Journal on Optimization, 20 0 (4): 0 1737--1765, 2010. doi:10.1137/080718206. URL https://doi.org/10.1137/080718206

    2010 doi

  22. [30]

    and Belkin, M

    Zhou, X. and Belkin, M. Semi-supervised learning by higher order regularization. In Gordon, G., Dunson, D., and Dudík, M. (eds.), Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, volume 15 of Proceedings of Machine Learning Rese...

    2011

  23. [31]

    Semi-supervised learning using Gaussian fields and harmonic functions

    Zhu, X., Ghahramani, Z., and Lafferty, J. Semi-supervised learning using Gaussian fields and harmonic functions. In Proceedings of the Twentieth International Conference on International Conference on Machine Learning, ICML'03, pp.\ 912–919. AAAI Press, 2003. ISBN 1577351894

    2003

  24. [32]

    Zhu, X. J. Semi-supervised learning literature survey. Technical report, University of Wisconsin-Madison Department of Computer Sciences, 2005

    2005

Pith tools

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