Pith. sign in

REVIEW 4 major objections 5 minor 32 references

Ambiguity-Guided Learnable Distribution Calibration for Semi-Supervised Few-Shot Class-Incremental Learning

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Pith's one-line read The paper redefines Semi-FSCIL so unlabeled data includes old and new classes, and proposes ALDC, a learnable threshold plus base-to-novel distribution calibration, which sets new state-of-the-art results on three benchmarks.

desk verdict The new GSemi-FSCIL setting is a real contribution, but the SOTA claim is undermined by an unmatched base-session protocol and an unclear calibration formula. read the letter →

arxiv 2507.23237 v1 pith:ZXRKDWK5 submitted 2025-07-31 cs.CV

classification cs.CV
keywords few-shotclass-incrementallearningsemi-supervisedgeneralizedsemi-FSCILdistributioncalibrationpseudo-labelingambiguity-guidedthreshold
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 argues that existing Semi-FSCIL settings are unrealistic because the unlabeled pool contains only current novel classes. It proposes Generalized Semi-FSCIL (GSemi-FSCIL), where unlabeled data also includes base and previously seen novel classes, and claims prior methods degrade in this setting. To address the resulting base-versus-novel confusion, it introduces ALDC: an ambiguity-guided learnable threshold selects which unlabeled samples to trust, and a base-to-novel calibration corrects biased few-shot class distributions using abundant base-class statistics. On mini-ImageNet, CIFAR-100, and CUB200, the paper reports state-of-the-art average accuracy, outperforming the prior Semi-FSCIL state of the art by 10.93% on mini-ImageNet.

What carries the argument

The machinery is the dual-branch Ambiguity-guided Learnable Threshold (ALT) combined with Base-to-Novel distribution calibration (B2N). ALT computes base and novel similarity scores $S_i^b$ and $S_i^n$ as maximum cosine similarities to the respective class weights, then sets a threshold $\tau = \frac{1}{N_u}\sum_i |S_i^b - S_i^n| + m$ to decide whether an unlabeled sample gets a pseudo-label or is treated as high-ambiguity. B2N estimates per-class mean $\mu_c$ and covariance $\Sigma_c$ for novel classes, averages them with the statistics of the most similar base classes via $\mu'_c = \frac{\sum_{i\in D_0}\mu_i + \mu_c}{N_t^c+1}$ and $\Sigma'_c = \frac{\sum_{i\in D_0}\Sigma_i + \Sigma_c}{N_t^c+1} + \alpha$, then samples synthetic features $x \sim \mathcal{N}(\mu'_c, \Sigma'_c)$ to augment the few-shot training set. This lets the classifier be retrained with an augmented set that the paper reports as converting an $N$-way $K$-shot task into an $N$-way $(K+10)$-shot task.

What would settle it

Retrain ALDC with the base-session protocol used by the baselines so that session-0 accuracy matches theirs, then recompute average accuracy over all sessions; if the 10.93% advantage shrinks or vanishes, the claimed improvement comes from initialization rather than from ALT and B2N.

Watch

Extended reading notes

Core claim

The central claim is that the main obstacle in generalized semi-supervised few-shot incremental learning is the presence of high-ambiguity unlabeled samples that resemble both base and novel classes, and that these samples can be exploited rather than discarded. The method computes per-sample similarity to base and novel class weights, uses a learnable threshold to separate confident samples from ambiguous ones, and for ambiguous samples calibrates the novel-class Gaussian feature distribution with base-class mean and covariance statistics, generating synthetic training features. The paper reports that this dual mechanism both protects base-class performance and improves novel-class accuracy, yielding new state-of-the-art results on three benchmarks.

Load-bearing premise

The headline gains assume the starting foundation is the same: ALDC enters at about 82% base-session accuracy while the methods it beats start near 72%, so if the base-session recipe is stronger, the improvement may come from initialization rather than from the new threshold and calibration modules.

Editorial extensions

If this is right

  • Prior Semi-FSCIL methods lose accuracy when the unlabeled pool includes base and previously seen novel classes, confirming that the generalized setting is genuinely harder.
  • Using a dynamic, learnable threshold beats both discarding ambiguous samples and using a fixed threshold, according to the paper's ablations.
  • Calibrating novel-class distributions with base-class statistics raises novel-class accuracy across sessions.
  • On mini-ImageNet, ALDC reaches 69.77% average accuracy, 10.93% above the best Semi-FSCIL method and 11.51% above the best GSemi-FSCIL baseline.

Reading between the lines

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

  • The GSemi-FSCIL protocol is a more faithful model of real deployments, where unlabeled streams mix old and new classes; if adopted as a standard evaluation, it could shift how semi-supervised incremental methods are compared.
  • Because B2N transfers covariance and mean from data-rich base classes to data-poor novel classes, the same mechanism is a natural fit for other label-scarce incremental settings where a related source class exists, such as class-incremental learning with pretrained features.
  • A direct stress test is to hold the base-session training recipe fixed across all methods, isolating whether the reported gain comes from ALT and B2N rather than from a stronger initialization.
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

4 major / 5 minor

Summary. The manuscript introduces a new setting, Generalized Semi-FSCIL (GSemi-FSCIL), in which the unlabeled pool in each incremental session contains samples from base and previously seen classes in addition to current novel classes. To handle this, the authors propose ALDC, consisting of an Ambiguity-guided Learnable Threshold (ALT) that pseudo-labels high-confidence unlabeled samples, and a Base-to-Novel (B2N) module that calibrates novel-class feature distributions using base-class statistics. The paper claims state-of-the-art results, with average accuracy gains of 10.93% over Semi-FSCIL and 11.51% over GSemi-FSCIL on mini-ImageNet, and presents curves for CIFAR-100 and CUB-200.

Significance. The proposed GSemi-FSCIL setting is a reasonable step toward more realistic semi-supervised FSCIL, and the observation that existing Semi-FSCIL methods degrade when the unlabeled set is generalized is potentially useful. If the reported gains survive a matched experimental protocol, ALDC would be a strong baseline for this setting. However, the paper's central SOTA claim is currently not established: the session-0 accuracy gap and the lack of a shared base-session training recipe dominate the reported average improvements, and the core B2N equation has formal problems. The paper does not include reproducible code or a parameter-free derivation, so the empirical comparison is the sole basis for the claims.

major comments (4)
  1. [Table I / Section V-A] The central SOTA claim is confounded by unequal session-0 conditions. ALDC reports 82.45% accuracy at session 0, while every baseline starts at 72.35% or lower; the reported average improvements of 10.93% and 11.51% are nearly the entire session-0 gap (10.10 points over UaD-CE). No base-session training protocol is given: Algorithm 1 only says 'Train backbone, classifier by D1', with no backbone, epochs, optimizer, learning rate, or augmentation details. Because the ablations in Tables II and III all build on the same 82.45% base model, they show only the marginal effect of ALT/B2N on that base, not that ALDC outperforms prior methods under matched initial conditions.
  2. [Eq. (4) / Section IV-B] The B2N calibration equation is dimensionally inconsistent as written. The numerator sums over i in D0 of class means (and covariances), but the denominator is N^t_c + 1, the per-class sample count plus one; if D0 consists of many base classes, the output is not a per-class mean or covariance estimate. The text says that only base classes estimated to be 'more similar' to the novel class should be used, but Eq. (4) pools all of D0. No derivation or normalization is supplied. Since B2N is one of the two core modules, this needs correction and validation.
  3. [Fig. 3(d) / Section V-D] The sensitivity analysis selects m and alpha by reporting the last-session accuracy, which appears to be test-set accuracy, and the manuscript never states which values of m and alpha are used in the main comparisons of Table I. This is a test-set model-selection issue that can bias the reported gains. The main-table numbers should be tied to a pre-specified validation choice, or the full sensitivity surface should be reported for all sessions rather than only session 8.
  4. [Fig. 3(a-b) / Section V-A] The three-dataset claim is not fully verifiable: CIFAR-100 and CUB-200 results are presented only as curves, with no numerical tables, no dataset-specific protocols, and no GSemi-FSCIL baseline numbers for those datasets. To support the claim of state-of-the-art results on three benchmark datasets, the paper should provide session-by-session accuracy tables and the same matched-protocol description for all datasets.
minor comments (5)
  1. [Section III] The definition D_t = {(x_i,y_i)}_{i=0}^{N times K} uses undefined notation, and the condition 't > 1' should be 't >= 1' for the first incremental session.
  2. [Algorithm 1] The symbol t is used both as the session index and as the threshold, which is confusing; the initial value of the threshold is never specified.
  3. [Eq. (2)] The smoothing coefficient m is not defined, and although a sensitivity grid is reported in Fig. 3(d), the manuscript does not state which m value was used for the main results.
  4. [Eq. (3)] The covariance formula sums over x_j in D_t, but it should sum over samples of class c; the meaning of N^t_c is not stated.
  5. [Tables II and III] The 'Baseline' average accuracy is 66.16 in Table II but 66.27 in Table III for identical per-session numbers; this inconsistency should be corrected.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: ALDC is an empirical pipeline whose SOTA claim rests on external benchmark comparison rather than deriving its results from its own definitions or self-citations.

full rationale

The paper's central claim is empirical: ALDC outperforms prior methods on three benchmark datasets. This claim is checked against held-out incremental sessions and external baselines, so it does not reduce to the paper's own definitions. The ALT threshold in Eq. (2) is a data-dependent heuristic for sample selection, not a fitted parameter renamed as a prediction. The B2N distribution calibration in Eq. (4) is an adaptation of the externally published distribution-calibration method of Yang et al. [30], which is independent of the present authors. The ablation studies measure the marginal contribution of ALT and B2N on top of a common base model, which is a legitimate internal comparison rather than a circular step. Some references are self-citations, but they appear only in related-work or acknowledgement contexts and are not load-bearing for the method, the setting, or the evaluation. The largest threat to the SOTA claim is the unmatched session-0 accuracy (ALDC starts at 82.45% while baselines start near 72.35% in Table I), which suggests a possible base-training-protocol confound; however, a comparison confound is an experimental-validity issue, not circularity, because the reported gains are not forced by construction or by the paper's equations. No step in the derivation chain is equivalent to its input by definition, and no prediction is statistically forced by a fitted parameter.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The central claim rests on a new task definition plus two heuristics. Free parameters (m, alpha, Nu) are tuned against the test benchmark, and Eq. (4) introduces an unproven transfer operation. No new physical or conceptual entities are postulated.

free parameters (3)
  • m (smoothing coefficient in Eq. 2) = tuned on {0, 0.2, 0.4, 0.6, 0.8, 1.0}
    Controls the threshold in the ambiguity-guided selection; selected based on final-session accuracy in Fig. 3(d).
  • alpha (covariance inflation in Eq. 4) = tuned on {0, 0.2, 0.4, 0.6, 0.8, 1.0}
    Added to the calibrated covariance; selected based on final-session accuracy in Fig. 3(d).
  • Nu (number of generated unlabeled samples per session) = 50, chosen from {25, 50, 75, 125}
    Table IV shows 125 samples gives higher average accuracy (70.80 vs 69.77), but 50 is used for the main comparison.
assumptions (3)
  • ad hoc to paper Base-class Gaussian statistics can be pooled and transferred to novel-class distributions via Eq. (4).
    No derivation is given; summing over all base-class means is not justified and appears dimensionally inconsistent.
  • domain assumption Cosine similarity between features and class weights is a sufficient signal for pseudo-labeling and ambiguity detection (Eqs. 1 and 2).
    Adopted as a standard heuristic from prior FSCIL work, without validation in the new mixed-unlabeled setting.
  • domain assumption Samples drawn from calibrated Gaussian distributions are valid training data for novel classes.
    Assumed in Section IV-B; no analysis is provided for the domain shift between generated and real feature vectors.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Ambiguity-Guided Learnable Distribution Calibration for Semi-Supervised Few-Shot Class-Incremental Learning." pith.science (2026). https://pith.science/paper/ZXRKDWK5

@misc{pith2026250723237,
  author       = {Pith},
  title        = {Pith review of: Ambiguity-Guided Learnable Distribution Calibration for Semi-Supervised Few-Shot Class-Incremental Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZXRKDWK5}},
  note         = {Machine review of arXiv:2507.23237}
}
read the original abstract

Few-Shot Class-Incremental Learning (FSCIL) focuses on models learning new concepts from limited data while retaining knowledge of previous classes. Recently, many studies have started to leverage unlabeled samples to assist models in learning from few-shot samples, giving rise to the field of Semi-supervised Few-shot Class-Incremental Learning (Semi-FSCIL). However, these studies often assume that the source of unlabeled data is only confined to novel classes of the current session, which presents a narrow perspective and cannot align well with practical scenarios. To better reflect real-world scenarios, we redefine Semi-FSCIL as Generalized Semi-FSCIL (GSemi-FSCIL) by incorporating both base and all the ever-seen novel classes in the unlabeled set. This change in the composition of unlabeled samples poses a new challenge for existing methods, as they struggle to distinguish between unlabeled samples from base and novel classes. To address this issue, we propose an Ambiguity-guided Learnable Distribution Calibration (ALDC) strategy. ALDC dynamically uses abundant base samples to correct biased feature distributions for few-shot novel classes. Experiments on three benchmark datasets show that our method outperforms existing works, setting new state-of-the-art results.

Figures

Figures reproduced from arXiv: 2507.23237 by the authors.

Figure 1
Figure 1. Comparisons between configurations of (a) FSCIL, (b) Semi-FSCIL, [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. An overview of our method for GSemi-FSCIL, where (a) and (b) illustrate the base and the incremental stages, respectively. Best viewed in color. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Performance curves of our method compared to -of-the-art FSCIL and Semi-FSCIL methods on (a) CIFAR100 and (b) CUB, (c) Performance of [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Examples of high-ambiguity samples. D. Visualization As shown in [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: The t-SNE visualization of the mini-ImageNet dataset is illustrated. [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 32 canonical work pages

  1. [1]

    Exploring example influence in continual learning,

    Q. Sun, F. Lyu, F. Shang, W. Feng, and L. Wan, “Exploring example influence in continual learning,” in Advances in Neural Information Processing Systems, 2022

  2. [2]

    Multi-domain multi-task rehearsal for lifelong learning,

    F. Lyu, S. Wang, W. Feng, Z. Ye, F. Hu, and S. Wang, “Multi-domain multi-task rehearsal for lifelong learning,” in Proceedings of the AAAI Conference on Artificial Intelligence , 2021

  3. [3]

    Harnessing multi-semantic hypergraph for few-shot learning,

    H. Chen, L. Li, Z. Xia, F. Lyu, L. Zhao, K. Huang, W. Feng, and F. Hu, “Harnessing multi-semantic hypergraph for few-shot learning,” in Chinese Conference on Pattern Recognition and Computer Vision , 2022

  4. [4]

    Measuring asymmetric gradient discrepancy in parallel continual learning,

    F. Lyu, Q. Sun, F. Shang, L. Wan, and W. Feng, “Measuring asymmetric gradient discrepancy in parallel continual learning,” Proceedings of the IEEE/CVF International Conference on Computer Vision , 2023

  5. [5]

    Multi-semantic hypergraph neural network for effective few- shot learning,

    H. Chen, L. Li, F. Hu, F. Lyu, L. Zhao, K. Huang, W. Feng, and Z. Xia, “Multi-semantic hypergraph neural network for effective few- shot learning,” Pattern Recognition, 2023

  6. [6]

    Metamask: Improving few- shot semantic segmentation via multi-mask calibration,

    D. Li, Z. Lu, W. Zheng, Q. Liao, and F. Lyu, “Metamask: Improving few- shot semantic segmentation via multi-mask calibration,” in International Joint Conference on Neural Networks , 2024

  7. [7]

    Safe: Slow and fast parameter-efficient tuning for continual learning with pre-trained models,

    L. Zhao, X. Zhang, K. Yan, S. Ding, and W. Huang, “Safe: Slow and fast parameter-efficient tuning for continual learning with pre-trained models,” Advances in Neural Information Processing Systems (NeurIPS), 2024

  8. [8]

    Few-shot class-incremental learning,

    X. Tao, X. Hong, X. Chang, S. Dong, X. Wei, and Y . Gong, “Few-shot class-incremental learning,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2020

Show all 32 references
  1. [9]

    Few-shot incre- mental learning with continually evolved classifiers,

    C. Zhang, N. Song, G. Lin, Y . Zheng, P. Pan, and Y . Xu, “Few-shot incre- mental learning with continually evolved classifiers,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021

  2. [10]

    For- ward compatible few-shot class-incremental learning,

    D.-W. Zhou, F.-Y . Wang, H.-J. Ye, L. Ma, S. Pu, and D.-C. Zhan, “For- ward compatible few-shot class-incremental learning,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022

  3. [11]

    Memorizing complementation network for few-shot class-incremental learning,

    Z. Ji, Z. Hou, X. Liu, Y . Pang, and X. Li, “Memorizing complementation network for few-shot class-incremental learning,” IEEE Transactions on Image Processing, 2023

  4. [12]

    Few- shot class-incremental learning via class-aware bilateral distillation,

    L. Zhao, J. Lu, Y . Xu, Z. Cheng, D. Guo, Y . Niu, and X. Fang, “Few- shot class-incremental learning via class-aware bilateral distillation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023

  5. [13]

    Cala: A class-aware logit adapter for few-shot class-incremental learning,

    C. Liu, L. Zhao, F. Lyu, K. Du, F. Hu, and T. Zhou, “Cala: A class-aware logit adapter for few-shot class-incremental learning,” arXiv preprint arXiv:2412.12654, 2024

  6. [14]

    A strong baseline for semi-supervised incremental few-shot learning,

    L. Zhao, D. Guo, Y . Xu, L. Qiao, Z. Cheng, S. Pu, Y . Niu, and X. Fang, “A strong baseline for semi-supervised incremental few-shot learning,” in British Machine Vision Conference (BMVC) , 2021

  7. [15]

    Uncertainty-aware distillation for semi-supervised few-shot class-incremental learning,

    Y . Cui, W. Deng, H. Chen, and L. Liu, “Uncertainty-aware distillation for semi-supervised few-shot class-incremental learning,” IEEE Trans- actions on Neural Networks and Learning Systems , 2023

  8. [16]

    Semi-supervised few- shot class-incremental learning,

    Y . Cui, W. Xiong, M. Tavakolian, and L. Liu, “Semi-supervised few- shot class-incremental learning,” in IEEE International Conference on Image Processing (ICIP) , 2021

  9. [17]

    Uncertainty-guided semi-supervised few-shot class-incremental learn- ing with knowledge distillation,

    Y . Cui, W. Deng, X. Xu, Z. Liu, Z. Liu, M. Pietik ¨ainen, and L. Liu, “Uncertainty-guided semi-supervised few-shot class-incremental learn- ing with knowledge distillation,” IEEE Transactions on Multimedia , 2024

  10. [18]

    Learning with fantasy: Semantic-aware virtual contrastive constraint for few-shot class- incremental learning,

    Z. Song, Y . Zhao, Y . Shi, P. Peng, L. Yuan, and Y . Tian, “Learning with fantasy: Semantic-aware virtual contrastive constraint for few-shot class- incremental learning,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2023

  11. [19]

    Few- shot class-incremental learning by sampling multi-phase tasks,

    D.-W. Zhou, H.-J. Ye, L. Ma, D. Xie, S. Pu, and D.-C. Zhan, “Few- shot class-incremental learning by sampling multi-phase tasks,” IEEE Transactions on Pattern Analysis and Machine Intelligence , 2022

  12. [20]

    Multi-feature space similarity supplement for few-shot class incremental learning,

    X. Xu, S. Niu, Z. Wang, W. Guo, L. Jing, and H. Yang, “Multi-feature space similarity supplement for few-shot class incremental learning,” Knowledge-Based Systems, 2023

  13. [21]

    Self-promoted prototype refinement for few-shot class-incremental learning,

    K. Zhu, Y . Cao, W. Zhai, J. Cheng, and Z.-J. Zha, “Self-promoted prototype refinement for few-shot class-incremental learning,” in Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021

  14. [22]

    Few-shot class-incremental learning via asymmetric supervised contrastive learn- ing,

    D. Liu, L. Zhao, Z. Zhang, F. Lyu, X. Fang, and L. Wang, “Few-shot class-incremental learning via asymmetric supervised contrastive learn- ing,” IEEE Transactions on Circuits and Systems for Video Technology , 2025

  15. [23]

    Virtual adversarial training: a regularization method for supervised and semi-supervised learning,

    T. Miyato, S.-i. Maeda, M. Koyama, and S. Ishii, “Virtual adversarial training: a regularization method for supervised and semi-supervised learning,” IEEE Transactions on Pattern Analysis and Machine Intelli- gence, vol. 41, no. 8, pp. 1979–1993, 2018

  16. [24]

    Mixmatch: A holistic approach to semi-supervised learning,

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

  17. [25]

    Transductive semi- supervised deep learning using min-max features,

    W. Shi, Y . Gong, C. Ding, Z. M. Tao, and N. Zheng, “Transductive semi- supervised deep learning using min-max features,” in Proceedings of the European conference on computer vision (ECCV) , 2018, pp. 299–315

  18. [26]

    Transductive learning via spectral graph partitioning,

    T. Joachims, “Transductive learning via spectral graph partitioning,” in Proceedings of the 20th International Conference on Machine Learning (ICML-03), 2003, pp. 290–297

  19. [27]

    Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks,

    D.-H. Lee et al., “Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks,” in Workshop on challenges in representation learning, ICML , vol. 3, no. 2. Atlanta, 2013, p. 896

  20. [28]

    Label propagation for deep semi-supervised learning,

    A. Iscen, G. Tolias, Y . Avrithis, and O. Chum, “Label propagation for deep semi-supervised learning,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, pp. 5070– 5079

  21. [29]

    Transductive inference for text classification using support vector machines,

    T. Joachims et al., “Transductive inference for text classification using support vector machines,” in Icml, vol. 99, 1999, pp. 200–209

  22. [30]

    Free lunch for few-shot learning: Distribution calibration,

    S. Yang, L. Liu, and M. Xu, “Free lunch for few-shot learning: Distribution calibration,” in International Conference on Learning Rep- resentations, 2021

  23. [31]

    icarl: Incremental classifier and representation learning,

    S.-A. Rebuffi, A. Kolesnikov, G. Sperl, and C. H. Lampert, “icarl: Incremental classifier and representation learning,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2017

  24. [32]

    Visualizing data using t-sne,

    L. van der Maaten and G. Hinton, “Visualizing data using t-sne,” Journal of Machine Learning Research , vol. 9, no. Nov, pp. 2579–2605, 2008

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.