Pith. sign in

REVIEW 3 major objections 7 minor 51 references

RegMixMatch: Optimizing Mixup Utilization in Semi-Supervised Learning

T0 review · 3 major / 7 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read RegMixMatch reports state-of-the-art semi-supervised error rates by keeping clean unlabeled samples alongside mixed ones and by class-aware mixing of low-confidence samples.

desk verdict Solid empirical SSL paper with strong low-label results; the CAM component is the weakest link but the overall method deserves serious review. read the letter →

arxiv 2412.10741 v2 pith:ETFSHERQ submitted 2024-12-14 cs.LG cs.CVstat.ML

Haorong Han , Jidong Yuan , Chixuan Wei , Zhongyang Yu This is my paper · ORCID
classification cs.LGcs.CVstat.ML
keywords semi-supervisedlearningMixuppseudo-labelingconsistencyregularizationlow-confidencesamplesclass-awaremixingartificiallabelpurityRegMixMatch
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 tries to establish that Mixup can be made effective in semi-supervised learning if it is used differently for high-confidence and low-confidence unlabeled examples. The authors argue that ordinary Mixup hurts SSL because interpolated labels are less pure, and that thresholding discards valuable low-confidence samples. Their RegMixMatch framework answers both problems: Semi-supervised RegMixup keeps clean unlabeled samples alongside mixed ones to preserve artificial-label purity, and Class-Aware Mixup blends low-confidence samples only with high-confidence samples sharing their predicted class, using top-2 class information and an $\ell^2$ loss. On standard benchmarks the method reports up to 4.35% error on CIFAR-10 with 10 labels and 11.74% on STL-10 with 40 labels, beating previous strong SSL methods in 9 of 12 settings. If these results hold, SSL can safely exploit samples below the pseudo-label threshold instead of discarding them.

What carries the argument

The machinery is a confidence split of the unlabeled batch into a high-confidence set $H$ and a low-confidence set $H^c$, each handled with its own loss. SRM uses pseudo-labels and cross-entropy for clean high-confidence samples, plus cross-entropy for ResizeMix-mixed high-confidence pairs; CAM uses softmax outputs, $\ell^2$ distance, and ResizeMix with a large mixing intensity $\alpha_l$ to mix low-confidence samples with same-class high-confidence partners. ResizeMix is the concrete image-and-label mixing operation, and thresholds $\tau_c$ and $\tau_m$ separate clean-sample supervision from mixing, with $\tau_m$ set higher to avoid early confirmation bias.

What would settle it

Run RegMixMatch with CAM replaced by class-unaware Mixup on a many-class dataset such as CIFAR-100 with 400 labels, while recording the top-2 accuracy of low-confidence unlabeled predictions; if that top-2 accuracy falls below the reliability of thresholded pseudo-labels, the CAM benefit should shrink or reverse, contradicting the paper's proposed mechanism.

Watch

Extended reading notes

Core claim

The central claim is that two standard reactions to Mixup in SSL are both suboptimal: using it everywhere as a data augmentation tool, or throwing away low-confidence samples. The paper's two components fix each. SRM adapts RegMixup's supervised recipe to SSL by training on clean high-confidence unlabeled samples with pseudo-labels and consistency regularization, while also training on mixed pairs of high-confidence samples, so Mixup acts as a regularizer rather than the main training signal. CAM then takes samples below the confidence threshold, uses their softmax output as the artificial label, and mixes each one with a high-confidence sample from the same predicted class using a stronger mixing intensity and an $\ell^2$ loss. The paper claims this converts noise in low-confidence artificial labels into useful information and reports that the combination reaches state-of-the-art errors in most low-label settings, including on ImageNet with a pre-trained backbone.

Load-bearing premise

The method's gains depend on the model's top-2 predicted class accuracy on unlabeled data staying high enough that low-confidence samples usually have their true class among the top two predictions; the paper demonstrates this during FixMatch training rather than within RegMixMatch's own optimization loop.

Editorial extensions

If this is right

  • RegMixMatch reports lower error than strong pseudo-labeling baselines in most low-label settings, so a hard confidence threshold need not be the final word on which unlabeled samples are useful.
  • Training on clean and mixed unlabeled samples together preserves artificial-label purity, which should let Mixup-based SSL recover the generalization gains Mixup provides in supervised learning.
  • Low-confidence samples, previously discarded, can be turned into training signal through class-aware mixing, improving accuracy per training iteration.
  • Stronger mixing intensities ($\alpha_h=1.0$, $\alpha_l=16.0$) become usable without hurting performance because clean samples are retained.
  • The recipe transfers beyond small benchmarks: on ImageNet with 10,000 labels it reports 3.66% lower error than a leading adaptive-threshold method.

Reading between the lines

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

  • Beyond the paper's experiments, the mechanism predicts that CAM's benefit tracks the model's top-2 accuracy on low-confidence data; on datasets with many classes or high inter-class similarity, that accuracy should drop and the CAM gain should shrink, which is a testable prediction.
  • Because the same-class partner is selected by predicted class, CAM also acts as a mild class-balancing mechanism; a natural extension would be to condition partner selection on class frequency to counter class imbalance in unlabeled data.
  • The high threshold $\tau_m=0.999$ suggests SRM behaves almost like clean-sample training in the early epochs; a curriculum that anneals $\tau_m$ downward as training progresses could extract even more signal from high-confidence mixing.
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 / 7 minor

Summary. RegMixMatch is a semi-supervised learning framework built from two Mixup-based mechanisms. Semi-supervised RegMixup (SRM) applies the RegMixup idea inside SSL: unlabeled samples whose prediction confidence exceeds threshold τc are pseudo-labeled and trained with weak-to-strong consistency (Eq. 7), while the subset above a higher threshold τm is further mixed with ResizeMix to give an additional mixed-sample cross-entropy term (Eq. 8); retaining the unmixed samples is intended to counteract the artificial-label purity loss that plain Mixup induces. Class-Aware Mixup (CAM) addresses samples below τm: each low-confidence strongly augmented sample is mixed with a high-confidence sample of the same predicted class and trained with an ℓ-2 loss on the soft-mixed label (Eq. 10), exploiting the observation that the true class of a low-confidence sample is usually within the top-2 predictions. The paper claims state-of-the-art error rates on 9 of 12 standard SSL settings (e.g., 4.35% on CIFAR-10 with 10 labels and 11.74% on STL-10 with 40 labels), plus ImageNet and USB pretrained-backbone results, ablations, a sensitivity analysis of τm, αh, and αl, and an efficiency comparison against FlatMatch, FreeMatch, and ReMixMatch.

Significance. The paper is a well-executed empirical contribution with several strengths: a clear diagnostic observation (Mixup degrades SSL by reducing artificial-label purity), a broad evaluation across 12 low-label settings plus ImageNet and the USB benchmark, standard deviations reported for the main results in the appendix, a released code repository, and an efficiency comparison. The SRM component is convincingly supported: removing its two terms costs 3.82 and 4.04 percentage points in the STL-10 40-label ablation (Table 3). The main weakness is that the second component, CAM, is supported only by (i) a motivating observation measured on FixMatch rather than on RegMixMatch (Figure 1b) and (ii) an ablation without variance whose effect size (0.38-0.56 points) is smaller than the method's reported standard deviation (0.57) on the same setting. If CAM's benefit is real, the combination is a strong new practical recipe for SSL; the evidence as presented does not yet establish it.

major comments (3)
  1. [Ablation Study, Table 3] Table 3 reports the STL-10 40-label ablation without standard deviations, and the two rows that isolate CAM (w/o CAM: 12.30, CAM→Mixup: 12.12) differ from the full method (11.74) by 0.56 and 0.38 percentage points. For the same setting, Table 10 reports the full method's standard deviation as ±0.57 and the FreeMatch baseline's as ±0.55, so the CAM effect is within roughly one standard deviation of the run-to-run noise and is not statistically distinguishable from zero as reported. Since the paper's central claim is that CAM, not just SRM, is responsible for the improvements, these ablation rows need to be rerun over multiple seeds and reported with variance, exactly as the main results in Table 10 were.
  2. [Introduction, Figure 1(b)] The motivating inequality — that top-2 predicted-class accuracy on unlabeled data exceeds pseudo-label reliability — is measured during FixMatch training, not during RegMixMatch training. SRM and CAM modify the training distribution, the pseudo-label thresholds, and the loss, so the inequality need not hold at the point where CAM is applied; the paper provides no on-RegMixMatch measurement of the same quantities, and Figure 4(b) is a single training trajectory without variance. Because the CAM mechanism is load-bearing for the paper's low-confidence-sample claims, the authors should report the reliability and top-2 accuracy curves for RegMixMatch itself, or provide seed-averaged learning curves for the CAM ablation.
  3. [Class-Aware Mixup, Eq. (10)] The class-aware pairing rule is underspecified. The loss requires, for each low-confidence sample i, a high-confidence partner j with arg max(q_j) = arg max(q_i), but the paper does not state what happens when no such partner exists in the current batch, whether the partner is drawn with or without replacement, or whether j = i is allowed. This is not a corner case: with the fixed threshold τm = 0.999, the high-confidence set H can be small or empty early in training, and in high-class-count settings such as ImageNet the same threshold may rarely be exceeded; this changes the effective Lcm term. The authors should state the exact sampling rule and report the typical size of H over training to make the algorithm reproducible and to substantiate that CAM is actually active in all claimed settings.
minor comments (7)
  1. [Experiments, Table 1] The main table omits standard deviations although they are reported in Table 10; given the large run-to-run variance in low-label settings (e.g., FreeMatch at 8.07±4.24 on CIFAR-10 with 10 labels), the headline margins should be accompanied by the corresponding variances, at least for the headline numbers.
  2. [Main Results] In the three second-best settings the method trails the best method by non-trivial margins on SVHN (1.77 vs 1.43 with 250 labels; 1.79 vs 1.41 with 1000 labels), so the claim of comprehensive superiority would benefit from a sentence discussing where and why RegMixMatch underperforms on SVHN.
  3. [Equations (7)-(11), Algorithm 1] The notation is confusing in places: the symbol H denotes both the cross-entropy loss and the high-confidence set, and in Eq. (10) the partner's pseudo-label q̂j is defined as arg max(q_i) rather than as the partner's own pseudo-label; in addition, Hc = U \ H includes samples with confidence between τc and τm, so calling all of Hc 'low-confidence' is imprecise.
  4. [Introduction, Figure 1(a) and Appendix Figure 6] The purity metric — the proportion of artificial labels whose highest predicted probability exceeds a threshold — never states which threshold is used; specify the threshold and the evaluation set for the curves in Figure 1(a) and Figure 6.
  5. [Efficiency Study, Figure 3] The scatter points appear to be single measurements per algorithm with no indication of the number of runs or the variance of the training-time measurement; state whether the times are per-iteration wall-clock averages over full training runs and over how many seeds.
  6. [Sensitivity Analysis, Figure 5] The sensitivity analysis is performed only on CIFAR-10 with 250 labels, and the resulting values (τm=0.999, αh=1.0, αl=16.0) are fixed across all other datasets including ImageNet; a sentence on how these values transfer, or on the sensitivity of the ImageNet results to τm given its 1000-class softmax, would strengthen the robustness claim.
  7. [Class-Aware Mixup] The sentence 'our experiments show that mixing more than two samples does not yield additional improvements' reports an experiment that is not shown anywhere in the paper or appendix; either present the experiment (e.g., in the appendix) or remove the claim.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: RegMixMatch's reported errors are external benchmark outcomes, and the only self-citation is non-load-bearing.

full rationale

The paper is an empirical SSL method rather than a derivation, and I could not find any step where a prediction reduces by construction to its input. SRM is explicitly built from the externally published RegMixup loss (Pinto et al. 2022), adapted to SSL with pseudo-labeling and consistency regularization; this is an application, not a renaming that silently imports the target result. CAM is a heuristic motivated by Figure 1(b), which measures top-2 accuracy during FixMatch training; that observation is not the paper's claim and is not used as the fitted target of the method. The only same-author citation, Wei et al. 2023, appears in a list of threshold-based pseudo-labeling works and is not load-bearing for any central argument. The ablation in Table 3 omits standard deviations and the CAM gain is small, but that is a statistical-reporting and robustness concern, not circularity. The headline error rates are external benchmark outcomes and are not fitted quantities, so there is no self-definitional or fitted-input-called-prediction pattern.

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

The framework rests on standard SSL distributional and consistency assumptions plus two author-defined heuristics: the purity metric and the top-2 accuracy observation. The main numerical choices (τm, αh, αl) are fitted on a single benchmark (CIFAR-10, 250 labels) through sensitivity analysis and then applied unchanged to all other datasets, which limits the independence of the reported SOTA results.

free parameters (5)
  • τm (high-confidence Mixup threshold) = 0.999
    Threshold above which pseudo-labels are used in Mixup; tuned on CIFAR-10 with 250 labels and fixed for all experiments (Section Sensitivity Analysis, Figure 5a).
  • αh (mixing intensity for high-confidence samples) = 1.0
    Beta distribution parameter controlling ResizeMix intensity; tuned on CIFAR-10 with 250 labels (Figure 5b).
  • αl (mixing intensity for low-confidence samples) = 16.0
    Beta distribution parameter controlling ResizeMix intensity for CAM; tuned on CIFAR-10 with 250 labels (Figure 5c).
  • τc (pseudo-label confidence threshold) = adaptive (FreeMatch) / 0.95 when fixed
    Uses FreeMatch's dynamic threshold with EMA decay 0.999; the fixed-threshold variant sets τc=0.95. Inherited from prior work but still a hyperparameter choice.
  • μ (unlabeled batch multiplier) = 7
    Standard SSL setting; number of unlabeled samples per labeled sample in a batch (Table 4).
assumptions (5)
  • domain assumption Unlabeled data is drawn from the same distribution as labeled data and shares the same label set in SSL benchmarks.
    Standard SSL assumption underlying pseudo-labeling and consistency loss (Section RegMixMatch).
  • domain assumption A model should output similar predictions for weakly and strongly augmented views of the same unlabeled image.
    The consistency regularization principle used in Equation 7.
  • domain assumption Pseudo-labels for samples whose confidence exceeds threshold τc are reliable enough to supervise the model.
    Assumed in the definition of Lu (Equation 7); if false, the clean-sample loss adds noise.
  • ad hoc to paper The purity metric, defined as the proportion of artificial labels with max probability above a threshold, is a meaningful proxy for label quality and learning progress.
    The metric is introduced by the authors to support the motivation and is not independently validated (Figure 1a, Section Introduction).
  • ad hoc to paper The top-2 accuracy of model predictions on unlabeled data exceeds the reliability of thresholded pseudo-labels, making class-aware mixing useful.
    Empirical observation from FixMatch training (Figure 1b) used to justify CAM; not re-evaluated for RegMixMatch.

how reviews work

0 comments
Cite this review

Pith. "Pith review of RegMixMatch: Optimizing Mixup Utilization in Semi-Supervised Learning." pith.science (2026). https://pith.science/paper/ETFSHERQ

@misc{pith2026241210741,
  author       = {Pith},
  title        = {Pith review of: RegMixMatch: Optimizing Mixup Utilization in Semi-Supervised Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ETFSHERQ}},
  note         = {Machine review of arXiv:2412.10741}
}
read the original abstract

Consistency regularization and pseudo-labeling have significantly advanced semi-supervised learning (SSL). Prior works have effectively employed Mixup for consistency regularization in SSL. However, our findings indicate that applying Mixup for consistency regularization may degrade SSL performance by compromising the purity of artificial labels. Moreover, most pseudo-labeling based methods utilize thresholding strategy to exclude low-confidence data, aiming to mitigate confirmation bias; however, this approach limits the utility of unlabeled samples. To address these challenges, we propose RegMixMatch, a novel framework that optimizes the use of Mixup with both high- and low-confidence samples in SSL. First, we introduce semi-supervised RegMixup, which effectively addresses reduced artificial labels purity by using both mixed samples and clean samples for training. Second, we develop a class-aware Mixup technique that integrates information from the top-2 predicted classes into low-confidence samples and their artificial labels, reducing the confirmation bias associated with these samples and enhancing their effective utilization. Experimental results demonstrate that RegMixMatch achieves state-of-the-art performance across various SSL benchmarks.

Figures

Figures reproduced from arXiv: 2412.10741 by the authors.

Figure 1
Figure 1. Motivating example of RegMixMatch. (a) Mixup’s [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overview of RegMixMatch. (a) shows the main idea of SRM. A weakly augmented image is fed into the model to [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Efficiency analysis of RegMixMatch. geNet with 10,000 labels. More comprehensive results and hyperparameter settings are detailed in the appendix. Efficiency Study For algorithms handling classification tasks, both runtime and classification performance are critical factors. In this section, we compare RegMixMatch with several SSL al￾gorithms in terms of both metrics. Specifically, we select FlatMatch (the previous … view at source ↗
Figures from the paper (3 more)
Figure 5
Figure 5. Figure 5: Parameter sensitivity analysis of RegMixMatch. [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Purity of different algorithms. 0 200k 400k 600k 800k 1000k Iter. 0.90 0.91 0.92 0.93 0.94 0.95 Accuracy FixMatch FreeMatch RegMixMatch [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]
Figure 7
Figure 7. Figure 7: Learning efficiency of different algorithms. [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

51 extracted references · 26 canonical work pages

  1. [1]

    Bachman, P.; Alsharif, O.; and Precup, D. 2014. Learning with pseudo-ensembles. Advances in neural information processing systems, 27

    work page 2014

  2. [2]

    Bengio, Y.; Louradour, J.; Collobert, R.; and Weston, J. 2009. Curriculum learning. In Proceedings of the 26th annual international conference on machine learning, 41--48

    2009

  3. [3]

    D.; Kurakin, A.; Sohn, K.; Zhang, H.; and Raffel, C

    Berthelot, D.; Carlini, N.; Cubuk, E. D.; Kurakin, A.; Sohn, K.; Zhang, H.; and Raffel, C. 2019 a . Remixmatch: Semi-supervised learning with distribution alignment and augmentation anchoring. arXiv preprint arXiv:1911.09785

    arXiv 2019

  4. [4]

    Berthelot, D.; Carlini, N.; Goodfellow, I.; Papernot, N.; Oliver, A.; and Raffel, C. A. 2019 b . Mixmatch: A holistic approach to semi-supervised learning. Advances in neural information processing systems, 32

    work page 2019

  5. [5]

    Chapelle, O.; Weston, J.; Bottou, L.; and Vapnik, V. 2000. Vicinal risk minimization. Advances in neural information processing systems, 13

    work page 2000

  6. [6]

    Chen, H.; Tao, R.; Fan, Y.; Wang, Y.; Wang, J.; Schiele, B.; Xie, X.; Raj, B.; and Savvides, M. 2023. SoftMatch: Addressing the quantity-quality tradeoff in semi-supervised learning. In The eleventh international conference on learning representations

    work page 2023

  7. [7]

    Chen, T.; Kornblith, S.; Norouzi, M.; and Hinton, G. 2020. A simple framework for contrastive learning of visual representations. In International conference on machine learning, 1597--1607. PMLR

    2020

  8. [8]

    Coates, A.; Ng, A.; and Lee, H. 2011. An analysis of single-layer networks in unsupervised feature learning. In Proceedings of the fourteenth international conference on artificial intelligence and statistics, 215--223

    work page 2011

Show all 51 references
  1. [9]

    D.; Zoph, B.; Mane, D.; Vasudevan, V.; and Le, Q

    Cubuk, E. D.; Zoph, B.; Mane, D.; Vasudevan, V.; and Le, Q. V. 2018. Autoaugment: Learning augmentation policies from data. arXiv preprint arXiv:1805.09501

    2018 arXiv

  2. [10]

    D.; Zoph, B.; Mane, D.; Vasudevan, V.; and Le, Q

    Cubuk, E. D.; Zoph, B.; Mane, D.; Vasudevan, V.; and Le, Q. V. 2019. Autoaugment: Learning augmentation strategies from data. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 113--123

    2019

  3. [11]

    D.; Zoph, B.; Shlens, J.; and Le, Q

    Cubuk, E. D.; Zoph, B.; Shlens, J.; and Le, Q. V. 2020. Randaugment: Practical automated data augmentation with a reduced search space. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops, 702--703

    2020

  4. [12]

    Deng, J.; Dong, W.; Socher, R.; Li, L.-J.; Li, K.; and Fei-Fei, L. 2009. Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, 248--255. Ieee

    2009

  5. [13]

    DeVries, T.; and Taylor, G. W. 2017. Improved regularization of convolutional neural networks with cutout. arXiv preprint arXiv:1708.04552

    2017 arXiv

  6. [14]

    Dosovitskiy, A. 2020. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929

    2020 arXiv

  7. [15]

    J.; Shlens, J.; and Szegedy, C

    Goodfellow, I. J.; Shlens, J.; and Szegedy, C. 2014. Explaining and harnessing adversarial examples. arXiv preprint arXiv:1412.6572

    2014 arXiv

  8. [16]

    He, K.; Chen, X.; Xie, S.; Li, Y.; Doll \'a r, P.; and Girshick, R. 2022. Masked autoencoders are scalable vision learners. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 16000--16009

    2022

  9. [17]

    He, K.; Zhang, X.; Ren, S.; and Sun, J. 2016. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, 770--778

    2016

  10. [18]

    Helber, P.; Bischke, B.; Dengel, A.; and Borth, D. 2019. Eurosat: A novel dataset and deep learning benchmark for land use and land cover classification. IEEE journal of selected topics in applied earth observations and remote sensing, 12(7): 2217--2226

    2019

  11. [19]

    Hong, M.; Choi, J.; and Kim, G. 2021. Stylemix: Separating content and style for enhanced data augmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 14862--14870

    2021

  12. [20]

    Huang, Z.; Shen, L.; Yu, J.; Han, B.; and Liu, T. 2023. Flatmatch: Bridging labeled data and unlabeled data with cross-sharpness for semi-supervised learning. Advances in neural information processing systems, 36: 18474--18494

    2023

  13. [21]

    Krizhevsky, A.; Hinton, G.; et al. 2009. Learning multiple layers of features from tiny images

    2009

  14. [22]

    Laine, S.; and Aila, T. 2017. Temporal ensembling for semi-supervised Learning. In International conference on learning representations

    2017

  15. [23]

    Lee, D.-H.; et al. 2013. Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks. In Workshop on challenges in representation learning, ICML, volume 3, 896. Atlanta

    2013

  16. [24]

    Li, J.; Xiong, C.; and Hoi, S. C. 2021. Comatch: Semi-supervised learning with contrastive graph regularization. In Proceedings of the IEEE/CVF international conference on computer vision, 9475--9484

    2021

  17. [25]

    Li, S.; Wang, Z.; Liu, Z.; Wu, D.; and Li, S. Z. 2023. Openmixup: Open mixup toolbox and benchmark for visual representation learning. arXiv preprint arXiv:2209.04851

    2023 arXiv

  18. [26]

    Liu, M.; Zeng, A.; Chen, M.; Xu, Z.; Lai, Q.; Ma, L.; and Xu, Q. 2022. Scinet: Time series modeling and forecasting with sample convolution and interaction. Advances in Neural Information Processing Systems, 35: 5816--5828

    2022

  19. [27]

    Miyato, T.; Maeda, S.-i.; Koyama, M.; and Ishii, S. 2018. Virtual adversarial training: a regularization method for supervised and semi-supervised learning. IEEE transactions on pattern analysis and machine intelligence, 41(8): 1979--1993

    2018

  20. [28]

    Y.; et al

    Netzer, Y.; Wang, T.; Coates, A.; Bissacco, A.; Wu, B.; Ng, A. Y.; et al. 2011. Reading digits in natural images with unsupervised feature learning. In NIPS workshop on deep learning and unsupervised feature learning, volume 2011, 4. Granada

    2011

  21. [29]

    Nguyen, K.-B. 2024. SequenceMatch: Revisiting the design of weak-strong augmentations for semi-supervised learning. In Proceedings of the IEEE/CVF winter conference on applications of computer vision, 96--106

    2024

  22. [30]

    Pham, H.; Dai, Z.; Xie, Q.; and Le, Q. V. 2021. Meta pseudo labels. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 11557--11568

    2021

  23. [31]

    N.; Torr, P.; and Dokania, P

    Pinto, F.; Yang, H.; Lim, S. N.; Torr, P.; and Dokania, P. 2022. Using mixup as a regularizer can surprisingly improve accuracy & out-of-distribution robustness. Advances in neural information processing systems, 35: 14608--14622

    2022

  24. [32]

    Qin, J.; Fang, J.; Zhang, Q.; Liu, W.; Wang, X.; and Wang, X. 2020. Resizemix: Mixing data with preserved object information and true labels. arXiv preprint arXiv:2012.11101

    2020 arXiv

  25. [33]

    Rasmus, A.; Berglund, M.; Honkala, M.; Valpola, H.; and Raiko, T. 2015. Semi-supervised learning with ladder networks. Advances in neural information processing systems, 28

    2015

  26. [34]

    A.; Cubuk, E

    Sohn, K.; Berthelot, D.; Carlini, N.; Zhang, Z.; Zhang, H.; Raffel, C. A.; Cubuk, E. D.; Kurakin, A.; and Li, C.-L. 2020. Fixmatch: Simplifying semi-supervised learning with consistency and confidence. Advances in neural information processing systems, 33: 596--608

    2020

  27. [35]

    Tarvainen, A.; and Valpola, H. 2017. Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results. Advances in neural information processing systems, 30

    2017

  28. [36]

    Uddin, A. F. M. S.; Monira, M. S.; Shin, W.; Chung, T.; and Bae, S.-H. 2021. SaliencyMix: A saliency guided data augmentation strategy for better regularization. In International conference on learning representations

    2021

  29. [37]

    Vapnik, V. 1991. Principles of risk minimization for learning theory. Advances in neural information processing systems, 4

    1991

  30. [38]

    Verma, V.; Lamb, A.; Beckham, C.; Najafi, A.; Mitliagkas, I.; Lopez-Paz, D.; and Bengio, Y. 2019. Manifold mixup: Better representations by interpolating hidden states. In International conference on machine learning, 6438--6447. PMLR

    2019

  31. [39]

    Wang, Y.; Chen, H.; Fan, Y.; Sun, W.; Tao, R.; Hou, W.; Wang, R.; Yang, L.; Zhou, Z.; Guo, L.-Z.; et al. 2022. Usb: A unified semi-supervised learning benchmark for classification. Advances in neural information processing systems, 35: 3938--3961

    2022

  32. [40]

    Wang, Y.; Chen, H.; Heng, Q.; Hou, W.; Fan, Y.; Wu, Z.; Wang, J.; Savvides, M.; Shinozaki, T.; Raj, B.; Schiele, B.; and Xie, X. 2023. FreeMatch: Self-adaptive thresholding for semi-supervised learning. In The eleventh international conference on learning representations

    2023

  33. [41]

    Wei, C.; Wang, Z.; Yuan, J.; Li, C.; and Chen, S. 2023. Time-frequency based multi-task learning for semi-supervised time series classification. Information sciences, 619: 762--780

    2023

  34. [42]

    W.; Snoek, J.; Lakshminarayanan, B.; and Tran, D

    Wen, Y.; Jerfel, G.; Muller, R.; Dusenberry, M. W.; Snoek, J.; Lakshminarayanan, B.; and Tran, D. 2021. Combining ensembles and data augmentation can harm your calibration. In International conference on learning representations

    2021

  35. [43]

    Xie, Q.; Dai, Z.; Hovy, E.; Luong, T.; and Le, Q. 2020. Unsupervised data augmentation for consistency training. Advances in neural information processing systems, 33: 6256--6268

    2020

  36. [44]

    Xu, Y.; Shang, L.; Ye, J.; Qian, Q.; Li, Y.-F.; Sun, B.; Li, H.; and Jin, R. 2021. Dash: Semi-supervised learning with dynamic thresholding. In International conference on machine learning, 11525--11536. PMLR

    2021

  37. [45]

    J.; Chun, S.; Choe, J.; and Yoo, Y

    Yun, S.; Han, D.; Oh, S. J.; Chun, S.; Choe, J.; and Yoo, Y. 2019. Cutmix: Regularization strategy to train strong classifiers with localizable features. In Proceedings of the IEEE/CVF international conference on computer vision, 6023--6032

    2019

  38. [46]

    Zagoruyko, S.; and Komodakis, N. 2016. Wide residual networks. arXiv preprint arXiv:1605.07146

    2016 arXiv

  39. [47]

    Zhang, B.; Wang, Y.; Hou, W.; Wu, H.; Wang, J.; Okumura, M.; and Shinozaki, T. 2021. Flexmatch: Boosting semi-supervised learning with curriculum pseudo labeling. Advances in neural information processing systems, 34: 18408--18419

    2021

  40. [48]

    N.; and Lopez-Paz, D

    Zhang, H.; Cisse, M.; Dauphin, Y. N.; and Lopez-Paz, D. 2018. mixup: Beyond empirical risk minimization. In International conference on learning representations

    2018

  41. [49]

    Zheng, M.; You, S.; Huang, L.; Wang, F.; Qian, C.; and Xu, C. 2022. Simmatch: Semi-supervised learning with similarity matching. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 14471--14481

    2022

  42. [50]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all...

  43. [51]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.