pith. sign in

arxiv: 2605.30135 · v1 · pith:HL4OHXBBnew · submitted 2026-05-28 · 💻 cs.LG · cs.AI

DAMEL: Dual-Axis Multi-Expert Learning for Class-Imbalanced Learning

Hyuck Lee , Taemin Park , Heeyoung Kim This is my paper

Pith reviewed 2026-06-29 08:38 UTC · model grok-4.3

classification 💻 cs.LG cs.AI
keywords class-imbalanced learningmulti-expert learningbias and variancelong-tailed distributionsrepresentation concatenationweight aggregationauxiliary classifier
0
0 comments X

The pith

DAMEL reduces both bias and variance in class-imbalanced learning by operating multiple experts along representation and time axes.

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

The paper introduces DAMEL as a multi-expert method that tackles class imbalance in long-tailed data without the usual bias-variance trade-off. It works along a representation axis by concatenating outputs from several experts and training an auxiliary balanced classifier on the combined features. Along a time axis it averages network weights from different training epochs and applies the averaged weights at test time. A reader would care because standard rebalancing methods cut bias but raise variance, while many multi-expert alternatives add procedural complexity. If the dual-axis design succeeds, it offers a comparatively direct route to lower error on real-world skewed datasets.

Core claim

DAMEL reduces both bias and variance of predictions by using multiple experts along both representation and time axes. Along the representation axis, DAMEL concatenates the representations of multiple experts and trains an auxiliary balanced classifier simultaneously with the concatenated representations. Along the time axis, DAMEL aggregates network weights across training epochs, employing these aggregated weights during testing.

What carries the argument

Dual-axis multi-expert learning: concatenation of expert representations plus an auxiliary balanced classifier on the representation axis, combined with epoch-wise weight aggregation on the time axis.

Load-bearing premise

That combining expert representations via concatenation plus an auxiliary balanced classifier and epoch-wise weight aggregation will simultaneously lower bias and variance without introducing offsetting increases in either or in computational cost.

What would settle it

On a standard long-tailed benchmark such as CIFAR-100-LT, run DAMEL against rebalancing baselines and multi-expert baselines; if measured bias or variance is not lower under DAMEL, the central claim does not hold.

Figures

Figures reproduced from arXiv: 2605.30135 by Heeyoung Kim, Hyuck Lee, Taemin Park.

Figure 1
Figure 1. Figure 1: Overall structure of DAMEL (two experts). Purple lines indicate forward [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Image representations taken differently by multiple experts. [PITH_FULL_IMAGE:figures/full_fig_p016_2.png] view at source ↗
read the original abstract

Various algorithms have been proposed to address the challenges posed by class-imbalanced learning from real-world data with long-tailed distributions. While these algorithms reduce prediction bias through rebalancing techniques, they often introduce increased prediction variance as a trade-off. Several multi-expert learning algorithms aim to address this variance but involve complex procedures. We propose a new multi-expert learning algorithm, called the dual-axis multi-expert learning (DAMEL), which reduces both bias and variance of predictions by using multiple experts along both representation and time axes. Along the representation axis, DAMEL concatenates the representations of multiple experts and trains an auxiliary balanced classifier simultaneously with the concatenated representations. Along the time axis, DAMEL aggregates network weights across training epochs, employing these aggregated weights during testing. Experimental results demonstrate that DAMEL reduces both bias and variance of predictions, highlighting its effectiveness in class-imbalanced learning.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

1 major / 1 minor

Summary. The manuscript proposes DAMEL, a dual-axis multi-expert learning algorithm for class-imbalanced (long-tailed) classification. Along the representation axis, multiple expert representations are concatenated and an auxiliary balanced classifier is trained jointly; along the time axis, network weights are aggregated across training epochs and the aggregated weights are used at test time. The central claim is that this approach simultaneously reduces both prediction bias and variance relative to prior rebalancing and multi-expert methods.

Significance. If the experimental results hold and the inference pipeline is clarified, DAMEL would supply a comparatively lightweight mechanism for addressing the bias-variance trade-off that commonly arises in imbalanced learning, without the complex procedures of some existing multi-expert ensembles.

major comments (1)
  1. [Abstract] Abstract (and §3, inference description): the test-time model is not fully specified. The abstract states that aggregated weights are used during testing but does not indicate whether the auxiliary balanced classifier or the concatenated multi-expert representation remains part of the inference pipeline. Because the bias-reduction claim is explicitly attributed to the representation-axis component, this omission is load-bearing for the dual-axis thesis.
minor comments (1)
  1. The abstract supplies no datasets, metrics, baselines, or quantitative results, which hinders immediate assessment of the strength of the experimental support.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive comment on clarifying the test-time inference pipeline. We address the point below.

read point-by-point responses

  1. Referee: [Abstract] Abstract (and §3, inference description): the test-time model is not fully specified. The abstract states that aggregated weights are used during testing but does not indicate whether the auxiliary balanced classifier or the concatenated multi-expert representation remains part of the inference pipeline. Because the bias-reduction claim is explicitly attributed to the representation-axis component, this omission is load-bearing for the dual-axis thesis.

    Authors: We agree this is a valid observation that affects clarity. The current abstract and §3 description of inference is incomplete regarding whether the concatenated representations and auxiliary balanced classifier are retained at test time. In the revised manuscript we will explicitly state that inference uses the aggregated weights applied to the concatenated multi-expert representations, which are then classified by the auxiliary balanced classifier. The same clarification will be added to §3. This change directly addresses the load-bearing aspect of the dual-axis claim. revision: yes

Circularity Check

0 steps flagged

No circularity; algorithmic proposal with experimental validation

full rationale

The paper proposes DAMEL as a new multi-expert algorithm that concatenates expert representations, trains an auxiliary balanced classifier, and aggregates weights over epochs. No equations, parameter fittings, derivations, or self-referential claims appear in the abstract or description. Central claims rest on experimental results rather than any reduction of outputs to inputs by construction. No self-citations are invoked as load-bearing premises. The method is self-contained as an empirical algorithmic contribution.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only review yields no explicit free parameters, invented entities, or non-standard axioms; the method implicitly relies on standard deep-learning assumptions such as the validity of SGD training and the utility of weight averaging.

axioms (1)
  • domain assumption Standard deep learning optimization assumptions (e.g., SGD convergence on neural networks) hold for the multi-expert training procedure.
    The method description presupposes that training multiple experts and aggregating weights behaves as expected under conventional optimizers.

pith-pipeline@v0.9.1-grok · 5680 in / 1203 out tokens · 31990 ms · 2026-06-29T08:38:11.134293+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

80 extracted references · 5 canonical work pages · 2 internal anchors

  1. [1]

    Why resampling outperforms reweight- ing for correcting sampling bias with stochastic gradients, in: Interna- tional Conference on Learning Representations

    An, J., Ying, L., Zhu, Y., 2021. Why resampling outperforms reweight- ing for correcting sampling bias with stochastic gradients, in: Interna- tional Conference on Learning Representations

    2021

  2. [2]

    Re- stricted decontamination for the imbalanced training sample problem, in: Iberoamerican Congress on Pattern Recognition, Springer

    Barandela, R., Rangel, E., Sánchez, J.S., Ferri, F.J., 2003. Re- stricted decontamination for the imbalanced training sample problem, in: Iberoamerican Congress on Pattern Recognition, Springer. pp. 424– 431

    2003

  3. [3]

    Weight uncertainty in neural network, in: International Conference on Machine Learning, PMLR

    Blundell, C., Cornebise, J., Kavukcuoglu, K., Wierstra, D., 2015. Weight uncertainty in neural network, in: International Conference on Machine Learning, PMLR. pp. 1613–1622

    2015

  4. [4]

    Ace: Ally complementary experts for solving long-tailed recognition in one-shot, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp

    Cai, J., Wang, Y., Hwang, J.N., 2021. Ace: Ally complementary experts for solving long-tailed recognition in one-shot, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 112–121

    2021

  5. [5]

    Learn- ing imbalanced datasets with label-distribution-aware margin loss, in: Wallach, H., Larochelle, H., Beygelzimer, A., d'Alché-Buc, F., Fox, E., Garnett, R

    Cao, K., Wei, C., Gaidon, A., Arechiga, N., Ma, T., 2019. Learn- ing imbalanced datasets with label-distribution-aware margin loss, in: Wallach, H., Larochelle, H., Beygelzimer, A., d'Alché-Buc, F., Fox, E., Garnett, R. (Eds.), Advances in Neural Information Processing Systems, Curran Associates, Inc

    2019

  6. [6]

    Smote: synthetic minority over-sampling technique

    Chawla, N.V., Bowyer, K.W., Hall, L.O., Kegelmeyer, W.P., 2002. Smote: synthetic minority over-sampling technique. Journal of Arti- ficial Intelligence Research 16, 321–357

    2002

  7. [7]

    Prediction of highly imbalanced semi- conductor chip-level defects in module tests using multimodal fusion and logit adjustment

    Cho, H., Koo, W., Kim, H., 2023. Prediction of highly imbalanced semi- conductor chip-level defects in module tests using multimodal fusion and logit adjustment. IEEE Transactions on Semiconductor Manufacturing 36, 425–433

    2023

  8. [8]

    Remix: Rebalanced mixup, in: European Conference on Computer Vision, Springer

    Chou, H.P., Chang, S.C., Pan, J.Y., Wei, W., Juan, D.C., 2020. Remix: Rebalanced mixup, in: European Conference on Computer Vision, Springer. pp. 95–110

    2020

  9. [9]

    Choy, M., Kim, D., Lee, J.G., Kim, H., Motoda, H., 2016. Looking back on the current day: interruptibility prediction using daily behavioral 25 features, in: Proceedings of the 2016 ACM international joint conference on pervasive and ubiquitous computing, pp. 1004–1015

    2016

  10. [10]

    Crime risk maps: A multivariate spatial analysis of crime data

    Chung, J., Kim, H., 2019. Crime risk maps: A multivariate spatial analysis of crime data. Geographical analysis 51, 475–499

    2019

  11. [11]

    Au- toaugment: Learning augmentation strategies from data, in: Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Cubuk, E.D., Zoph, B., Mane, D., Vasudevan, V., Le, Q.V., 2019. Au- toaugment: Learning augmentation strategies from data, in: Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 113–123

    2019

  12. [12]

    Reslt: Residual learn- ing for long-tailed recognition

    Cui, J., Liu, S., Tian, Z., Zhong, Z., Jia, J., 2022. Reslt: Residual learn- ing for long-tailed recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence 45, 3695–3706

    2022

  13. [13]

    Parametric contrastive learning, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp

    Cui, J., Zhong, Z., Liu, S., Yu, B., Jia, J., 2021. Parametric contrastive learning, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 715–724

    2021

  14. [14]

    Class- balanced loss based on effective number of samples, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recog- nition, pp

    Cui, Y., Jia, M., Lin, T.Y., Song, Y., Belongie, S., 2019. Class- balanced loss based on effective number of samples, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recog- nition, pp. 9268–9277

    2019

  15. [15]

    On supervised class-imbalanced learning: An updated perspective and some key challenges

    Das, S., Mullick, S.S., Zelinka, I., 2022. On supervised class-imbalanced learning: An updated perspective and some key challenges. IEEE Trans- actions on Artificial Intelligence 3, 973–993

    2022

  16. [16]

    Ima- genet: A large-scale hierarchical image database, in: 2009 IEEE confer- ence on computer vision and pattern recognition, Ieee

    Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L., 2009. Ima- genet: A large-scale hierarchical image database, in: 2009 IEEE confer- ence on computer vision and pattern recognition, Ieee. pp. 248–255

    2009

  17. [17]

    Improved Regularization of Convolutional Neural Networks with Cutout

    DeVries, T., Taylor, G.W., 2017. Improved regularization of convolu- tional neural networks with cutout. arXiv preprint arXiv:1708.04552

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  18. [18]

    Probabilistic contrastive learningforlong-tailedvisualrecognition

    Du, C., Wang, Y., Song, S., Huang, G., 2024. Probabilistic contrastive learningforlong-tailedvisualrecognition. IEEETransactionsonPattern Analysis and Machine Intelligence 46, 5890–5904. 26

    2024

  19. [19]

    Fan, Y., Dai, D., Kukleva, A., Schiele, B., 2022. Cossl: Co-learning of representation and classifier for imbalanced semi-supervised learning, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 14574–14584

    2022

  20. [20]

    Deep Ensem- bles: A Loss Landscape Perspective,

    Fort, S., Hu, H., Lakshminarayanan, B., 2019. Deep ensembles: A loss landscape perspective. arXiv preprint arXiv:1912.02757

    work page arXiv 2019

  21. [21]

    Learning from imbalanced data

    He, H., Garcia, E.A., 2009. Learning from imbalanced data. IEEE Transactions on knowledge and data engineering 21, 1263–1284

    2009

  22. [22]

    Momentum contrast for unsupervised visual representation learning, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    He, K., Fan, H., Wu, Y., Xie, S., Girshick, R., 2020. Momentum contrast for unsupervised visual representation learning, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 9729–9738

    2020

  23. [23]

    Deep residual learning for image recognition, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp

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

    2016

  24. [24]

    Searching for mo- bilenetv3, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp

    Howard, A., Sandler, M., Chu, G., Chen, L.C., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., Vasudevan, V., et al., 2019. Searching for mo- bilenetv3, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 1314–1324

    2019

  25. [25]

    Learning deep representa- tion for imbalanced classification, in: Proceedings of the IEEE confer- ence on computer vision and pattern recognition, pp

    Huang, C., Li, Y., Loy, C.C., Tang, X., 2016. Learning deep representa- tion for imbalanced classification, in: Proceedings of the IEEE confer- ence on computer vision and pattern recognition, pp. 5375–5384

    2016

  26. [26]

    Class-balanced distil- lation for long-tailed visual recognition

    Iscen, A., Araujo, A., Gong, B., Schmid, C., 2021. Class-balanced distil- lation for long-tailed visual recognition. arXiv preprint arXiv:2104.05279

    work page arXiv 2021

  27. [27]

    Izmailov, P., Podoprikhin, D., Garipov, T., Vetrov, D., Wilson, A.G.,

  28. [28]

    Averaging weights leads to wider optima and better generalization, in: 34th Conference on Uncertainty in Artificial Intelligence 2018, UAI 2018, Association For Uncertainty in Artificial Intelligence (AUAI). pp. 876–885

    2018

  29. [29]

    Jamal, M.A., Brown, M., Yang, M.H., Wang, L., Gong, B., 2020. Re- thinking class-balanced methods for long-tailed visual recognition from a 27 domain adaptation perspective, in: Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pp. 7610–7619

    2020

  30. [30]

    The class imbalance problem: Significance and strategies, in: Proc

    JAPKOWICZ, N., 2000. The class imbalance problem: Significance and strategies, in: Proc. 2000 International Conference on Artificial Intelligence, pp. 111–117

    2000

  31. [31]

    Exploring balanced feature spaces for representation learning, in: International Conference on Learning Representations

    Kang, B., Li, Y., Xie, S., Yuan, Z., Feng, J., 2020a. Exploring balanced feature spaces for representation learning, in: International Conference on Learning Representations

  32. [32]

    Decoupling representation and classifier for long-tailed recognition, in: International Conference on Learning Representations

    Kang, B., Xie, S., Rohrbach, M., Yan, Z., Gordo, A., Feng, J., Kalan- tidis, Y., 2020b. Decoupling representation and classifier for long-tailed recognition, in: International Conference on Learning Representations

  33. [33]

    Exploiting shape cues for weakly supervised semantic segmentation

    Kho, S., Lee, P., Lee, W., Ki, M., Byun, H., 2022. Exploiting shape cues for weakly supervised semantic segmentation. Pattern Recognition 132, 108953

    2022

  34. [34]

    Contextual anomaly detection for high- dimensional data using dirichlet process variational autoencoder

    Kim, H., Kim, H., 2023. Contextual anomaly detection for high- dimensional data using dirichlet process variational autoencoder. IISE Transactions 55, 433–444

    2023

  35. [35]

    M2m: Imbalanced classification via major-to-minor translation, in: Proceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition, pp

    Kim, J., Jeong, J., Shin, J., 2020. M2m: Imbalanced classification via major-to-minor translation, in: Proceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition, pp. 13896–13905

    2020

  36. [36]

    Locally most powerful bayesian test for out-of-distribution detection using deep generative models

    Kim, K., Shin, J., Kim, H., 2021. Locally most powerful bayesian test for out-of-distribution detection using deep generative models. Advances in Neural Information Processing Systems 34, 14913–14924

    2021

  37. [37]

    Flexible model weighting for one-dependence estimators based on point-wise independence analysis

    Kong, H., Wang, L., 2023. Flexible model weighting for one-dependence estimators based on point-wise independence analysis. Pattern Recog- nition 139, 109473

    2023

  38. [38]

    Learning multiple layers of features from tiny images

    Krizhevsky, A., 2009. Learning multiple layers of features from tiny images. Technical report, Department of Computer Science, University of Toronto

    2009

  39. [39]

    Simple and scal- able predictive uncertainty estimation using deep ensembles

    Lakshminarayanan, B., Pritzel, A., Blundell, C., 2017. Simple and scal- able predictive uncertainty estimation using deep ensembles. Advances in Neural Information Processing Systems 30. 28

    2017

  40. [40]

    Semi-supervised learning for simulta- neous location detection and classification of mixed-type defect patterns in wafer bin maps

    Lee, H., Lee, J., Kim, H., 2023a. Semi-supervised learning for simulta- neous location detection and classification of mixed-type defect patterns in wafer bin maps. IEEE Transactions on Semiconductor Manufacturing 36, 220–230

  41. [41]

    Lee, H., Park, T., Kim, H., 2025. Learnable logit adjustment for imbal- anced semi-supervised learning under class distribution mismatch, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 2664–2674

    2025

  42. [42]

    Abc: Auxiliary balanced classifier for class-imbalanced semi-supervised learning

    Lee, H., Shin, S., Kim, H., 2021. Abc: Auxiliary balanced classifier for class-imbalanced semi-supervised learning. Advances in Neural Informa- tion Processing Systems 34

    2021

  43. [43]

    Resampling approach for one-class classification

    Lee, H.H., Park, S., Im, J., 2023b. Resampling approach for one-class classification. Pattern Recognition , 109731

  44. [44]

    Dependence maps, a dimensionality reduction with dependence distance for high-dimensional data

    Lee, K., Gray, A., Kim, H., 2013. Dependence maps, a dimensionality reduction with dependence distance for high-dimensional data. Data Mining and Knowledge Discovery 26, 512–532

    2013

  45. [45]

    Trustworthy long- tailed classification, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Li, B., Han, Z., Li, H., Fu, H., Zhang, C., 2022a. Trustworthy long- tailed classification, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 6970–6979

  46. [46]

    Targeted supervised contrastive learning for long- tailed recognition, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Li, T., Cao, P., Yuan, Y., Fan, L., Yang, Y., Feris, R.S., Indyk, P., Katabi, D., 2022b. Targeted supervised contrastive learning for long- tailed recognition, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 6918–6928

  47. [47]

    Self supervision to distillation for long-tailed visual recognition, in: Proceedings of the IEEE/CVF Inter- national Conference on Computer Vision, pp

    Li, T., Wang, L., Wu, G., 2021. Self supervision to distillation for long-tailed visual recognition, in: Proceedings of the IEEE/CVF Inter- national Conference on Computer Vision, pp. 630–639

    2021

  48. [48]

    Large- scale long-tailed recognition in an open world, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Liu, Z., Miao, Z., Zhan, X., Wang, J., Gong, B., Yu, S.X., 2019. Large- scale long-tailed recognition in an open world, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 2537–2546. 29

    2019

  49. [49]

    A high-bias, low-variance introduction to machine learning for physicists

    Mehta, P., Bukov, M., Wang, C.H., Day, A.G., Richardson, C., Fisher, C.K., Schwab, D.J., 2019. A high-bias, low-variance introduction to machine learning for physicists. Physics reports 810, 1–124

    2019

  50. [50]

    Long-tail learning via logit adjustment, in: International Con- ference on Learning Representations

    Menon, A.K., Jayasumana, S., Rawat, A.S., Jain, H., Veit, A., Kumar, S., 2020. Long-tail learning via logit adjustment, in: International Con- ference on Learning Representations

    2020

  51. [51]

    Distributed representations of words and phrases and their composi- tionality, in: Burges, C.J.C., Bottou, L., Welling, M., Ghahramani, Z., Weinberger, K.Q

    Mikolov, T., Sutskever, I., Chen, K., Corrado, G.S., Dean, J., 2013. Distributed representations of words and phrases and their composi- tionality, in: Burges, C.J.C., Bottou, L., Welling, M., Ghahramani, Z., Weinberger, K.Q. (Eds.), Advances in Neural Information Processing Systems, Curran Associates, Inc

    2013

  52. [52]

    Eigen-cam: Class activation map using principal components, in: 2020 International Joint Conference on Neural Networks (IJCNN), IEEE

    Muhammad, M.B., Yeasin, M., 2020. Eigen-cam: Class activation map using principal components, in: 2020 International Joint Conference on Neural Networks (IJCNN), IEEE. pp. 1–7

    2020

  53. [53]

    Diversity matters when learning from ensembles

    Nam, G., Yoon, J., Lee, Y., Lee, J., 2021. Diversity matters when learning from ensembles. Advances in Neural Information Processing Systems 34

    2021

  54. [54]

    Prediction of highly imbalanced semi- conductor chip-level defects using uncertainty-based adaptive margin learning

    Park, S., Kim, K., Kim, H., 2022. Prediction of highly imbalanced semi- conductor chip-level defects using uncertainty-based adaptive margin learning. IISE Transactions 55, 147–155

    2022

  55. [55]

    Rebalancing using estimated class dis- tribution for imbalanced semi-supervised learning under class distribu- tion mismatch, in: European Conference on Computer Vision, Springer

    Park, T., Lee, H., Kim, H., 2024. Rebalancing using estimated class dis- tribution for imbalanced semi-supervised learning under class distribu- tion mismatch, in: European Conference on Computer Vision, Springer. pp. 388–404

    2024

  56. [56]

    Bal- anced meta-softmax for long-tailed visual recognition, in: Larochelle, H., Ranzato, M., Hadsell, R., Balcan, M.F., Lin, H

    Ren, J., Yu, C., sheng, s., Ma, X., Zhao, H., Yi, S., Li, h., 2020. Bal- anced meta-softmax for long-tailed visual recognition, in: Larochelle, H., Ranzato, M., Hadsell, R., Balcan, M.F., Lin, H. (Eds.), Advances in Neural Information Processing Systems, Curran Associates, Inc.. pp. 4175–4186

    2020

  57. [57]

    Learning to reweight examples for robust deep learning, in: International Conference on Ma- chine Learning, PMLR

    Ren, M., Zeng, W., Yang, B., Urtasun, R., 2018. Learning to reweight examples for robust deep learning, in: International Conference on Ma- chine Learning, PMLR. pp. 4334–4343. 30

    2018

  58. [58]

    How re-sampling helps for long-tail learning?, in: Advances in Neural Information Processing Systems

    Shi, J.X., Wei, T., Xiang, Y., Li, Y.F., 2023. How re-sampling helps for long-tail learning?, in: Advances in Neural Information Processing Systems

    2023

  59. [59]

    Application of kernel principal component analysis to multi-characteristic parameter design problems

    Soh, W., Kim, H., Yum, B.J., 2018. Application of kernel principal component analysis to multi-characteristic parameter design problems. Annals of Operations Research 263, 69–91

    2018

  60. [60]

    Mnasnet: Platform-aware neural architecture search for mobile, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Tan, M., Chen, B., Pang, R., Vasudevan, V., Sandler, M., Howard, A., Le, Q.V., 2019. Mnasnet: Platform-aware neural architecture search for mobile, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 2820–2828

    2019

  61. [61]

    Efficientnet: Rethinking model scaling for con- volutional neural networks, in: International Conference on Machine Learning, PMLR

    Tan, M., Le, Q., 2019. Efficientnet: Rethinking model scaling for con- volutional neural networks, in: International Conference on Machine Learning, PMLR. pp. 6105–6114

    2019

  62. [62]

    Tarvainen, A., Valpola, H., 2017. Mean teachers are better role mod- els: Weight-averaged consistency targets improve semi-supervised deep learning results, in: Proceedings of the 31st International Conference on Neural Information Processing Systems, pp. 1195–1204

    2017

  63. [63]

    The inaturalist species classi- fication and detection dataset, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp

    Van Horn, G., Mac Aodha, O., Song, Y., Cui, Y., Sun, C., Shepard, A., Adam, H., Perona, P., Belongie, S., 2018. The inaturalist species classi- fication and detection dataset, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 8769–8778

    2018

  64. [64]

    Contrastive learning based hybrid networks for long-tailed image classification, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, pp

    Wang, P., Han, K., Wei, X.S., Zhang, L., Wang, L., 2021a. Contrastive learning based hybrid networks for long-tailed image classification, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, pp. 943–952

  65. [65]

    Long-tailed recog- nition by routing diverse distribution-aware experts, in: International Conference on Learning Representations

    Wang, X., Lian, L., Miao, Z., Liu, Z., Yu, S., 2021b. Long-tailed recog- nition by routing diverse distribution-aware experts, in: International Conference on Learning Representations

  66. [66]

    Multiple contrastive experts for long-tailed image classification

    Wang, Y., Sun, K., Guo, C., Zhong, S., Liu, H., Ma, Y., 2024. Multiple contrastive experts for long-tailed image classification. Expert Systems with Applications 255, 124613. 31

    2024

  67. [67]

    Learning to model the tail, in: Guyon, I., Luxburg, U.V., Bengio, S., Wallach, H., Fergus, R., Vishwanathan, S., Garnett, R

    Wang, Y.X., Ramanan, D., Hebert, M., 2017. Learning to model the tail, in: Guyon, I., Luxburg, U.V., Bengio, S., Wallach, H., Fergus, R., Vishwanathan, S., Garnett, R. (Eds.), Advances in Neural Information Processing Systems, Curran Associates, Inc

    2017

  68. [68]

    Open-sampling: Exploring out-of-distribution data for re-balancing long-tailed datasets, in: International Conference on Machine Learning, PMLR

    Wei, H., Tao, L., Xie, R., Feng, L., An, B., 2022. Open-sampling: Exploring out-of-distribution data for re-balancing long-tailed datasets, in: International Conference on Machine Learning, PMLR. pp. 23615– 23630

    2022

  69. [69]

    Flipout: Efficient pseudo-independent weight perturbations on mini-batches, in: Interna- tional Conference on Learning Representations

    Wen, Y., Vicol, P., Ba, J., Tran, D., Grosse, R., 2018. Flipout: Efficient pseudo-independent weight perturbations on mini-batches, in: Interna- tional Conference on Learning Representations

    2018

  70. [70]

    Learning from multiple experts: Self- paced knowledge distillation for long-tailed classification, in: European Conference on Computer Vision, Springer

    Xiang, L., Ding, G., Han, J., 2020. Learning from multiple experts: Self- paced knowledge distillation for long-tailed classification, in: European Conference on Computer Vision, Springer. pp. 247–263

    2020

  71. [71]

    Aggregated residual transformations for deep neural networks, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp

    Xie, S., Girshick, R., Dollár, P., Tu, Z., He, K., 2017. Aggregated residual transformations for deep neural networks, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1492–1500

    2017

  72. [72]

    Feature transfer learning for deep face recognition with under-represented data

    Yin, X., Yu, X., Sohn, K., Liu, X., Chandraker, M., 2018. Feature transfer learning for deep face recognition with under-represented data. arXiv e-prints , arXiv–1803

    2018

  73. [73]

    Multimodal deep generative model for semi- supervised learning under class imbalance

    Yoon, H., Kim, H., 2026. Multimodal deep generative model for semi- supervised learning under class imbalance. Technometrics , 1–24

    2026

  74. [74]

    Uncertainty estimation by density aware evi- dential deep learning

    Yoon, T., Kim, H., 2024. Uncertainty estimation by density aware evi- dential deep learning. arXiv preprint arXiv:2409.08754

    work page arXiv 2024

  75. [75]

    Pure noise to the rescue of insuffi- cient data: Improving imbalanced classification by training on random noiseimages, in: InternationalConferenceonMachineLearning, PMLR

    Zada, S., Benou, I., Irani, M., 2022. Pure noise to the rescue of insuffi- cient data: Improving imbalanced classification by training on random noiseimages, in: InternationalConferenceonMachineLearning, PMLR. pp. 25817–25833

    2022

  76. [76]

    mixup: Beyond Empirical Risk Minimization

    Zhang, H., Cisse, M., Dauphin, Y.N., Lopez-Paz, D., 2017. mixup: Beyond empirical risk minimization. arXiv preprint arXiv:1710.09412 . 32

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  77. [77]

    mixup: Beyond empirical risk minimization, in: International Conference on Learning Representations

    Zhang, H., Cisse, M., Dauphin, Y.N., Lopez-Paz, D., 2018. mixup: Beyond empirical risk minimization, in: International Conference on Learning Representations. URL:https://openreview.net/forum?id= r1Ddp1-Rb

    2018

  78. [78]

    Siamese networks with an online reweighted example for imbalanced data learning

    Zhao, L., Shang, Z., Tan, J., Zhou, M., Zhang, M., Gu, D., Zhang, T., Tang, Y.Y., 2022. Siamese networks with an online reweighted example for imbalanced data learning. Pattern Recognition 132, 108947

    2022

  79. [79]

    Zhou, B., Cui, Q., Wei, X.S., Chen, Z.M., 2020. Bbn: Bilateral-branch network with cumulative learning for long-tailed visual recognition, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, pp. 9719–9728

    2020

  80. [80]

    Class and attribute-aware logit adjustment for generalized long-tail learning, in: Proceedings of the AAAI Conference on Artificial Intelligence

    Zhou, X., Wu, O., Yang, N., 2025. Class and attribute-aware logit adjustment for generalized long-tail learning, in: Proceedings of the AAAI Conference on Artificial Intelligence. 33

    2025