Pith. sign in

REVIEW 4 major objections 5 minor 96 references

Maximally Separated Active Learning

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Using fixed equiangular class prototypes as the network head, this paper shows that active learning gets both uncertainty and diversity without clustering, outperforming several baselines across five image benchmarks.

desk verdict Simple repurposing of fixed hyperspherical prototypes for active learning; the core idea is real, but the 'wins across five datasets' claim overstates the tables and the diversity step is untested where it should hurt most. read the letter →

arxiv 2411.17444 v1 pith:PZ6CHPLU submitted 2024-11-26 cs.LG

Tejaswi Kasarla , Abhishek Jha , Faye Tervoort , Rita Cucchiara , Pascal Mettes This is my paper · ORCID
classification cs.LG
keywords activelearninghypersphericalprototypesuncertaintysamplingdiversitymaximumseparationimageclassificationlong-tailedpseudo-labeling
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

Active learning chooses which unlabeled examples to label next, typically by picking the model's most uncertain samples, but that tends to select similar, redundant examples. This paper argues that fixing a set of maximally separated (equiangular) points on the hypersphere as class prototypes solves both sides of the problem at once: the prototypes act as an inductive bias that stabilizes class representations, and cosine similarity to the nearest prototype doubles as an uncertainty score. To add diversity without clustering, the method pre-filters the most uncertain examples and then takes an equal number of those closest to each prototype, treating prototypes as cluster centers. The paper reports that this strategy, MSAL-D, achieves the best overall accuracy-budget scores on MNIST, SVHN, and TinyImageNet, and that adding the fixed prototype head improves several classic uncertainty metrics on most settings. A sympathetic reader would take the central claim to be that hyperspherical uniformity can replace separate clustering and representation-learning machinery in active learning.

What carries the argument

The carrying object is the fixed equiangular prototype matrix $P_c \in \mathbb{R}^{(C-1) \times C}$ built by the maximum-separation construction (the recursive formula with vectors separated by angle $-1/k$), used as the network's fixed last layer so that logits are $\rho P_c^T \Phi(x)$. The uncertainty score $\alpha_i^{\mathrm{MSAL}} = \max_c ( \hat{x}_i \cdot P_c^T ) / \max \| \hat{x}_i \|_2$ measures how far a sample sits from every prototype; the diversity step treats the $C$ prototypes as permanent cluster centers and allocates the batch by taking $b/C$ closest samples per prototype from the pre-filtered $\beta b$ most uncertain samples. The paper's key mechanism is that a single fixed geometry does both jobs: it shapes the learned embedding space through training and it partitions the uncertain pool for batch selection.

What would settle it

The paper's own long-tailed results provide a check: on CIFAR-100 with imbalance factor 0.1, MSAL's AUBC of 0.3230 falls below margin sampling's 0.3263. A controlled run that varies the per-prototype allocation from equal ($b/C$) to class-frequency-proportional, keeping all other settings fixed, would settle whether the equal-allocation diversity heuristic is what hurts under imbalance and whether a simple correction restores the claimed edge.

Watch

Extended reading notes

Core claim

Writing as the authors would: enforcing maximum separation through fixed class prototypes in the network's last layer yields stable, well-separated embeddings from the first active learning round onward, and those same prototypes give a simple, cluster-free way to select both uncertain and diverse samples. Uncertainty is scored as the cosine similarity between a sample's feature and its closest prototype; samples far from all prototypes are most uncertain. Diversity is obtained by retaining the $\beta b$ most uncertain samples, then selecting the $b/C$ samples closest to each of the $C$ prototypes, so the queried batch spreads across the class-separated regions of the space. On five image benchmarks (MNIST, SVHN, CIFAR-10, CIFAR-100, TinyImageNet) the combined strategy MSAL-D reports the highest area-under-the-budget-curve (AUBC) on three of them, and swapping in other uncertainty metrics (least confidence, margin, entropy) under the fixed-prototype head instead of a trainable head generally improves their results.

Load-bearing premise

The diversity step assumes that the learned features are calibrated around the fixed prototypes, so partitioning the pre-filtered uncertain pool by nearest prototype and taking $b/C$ samples per class yields a representative and diverse batch.

Editorial extensions

If this is right

  • MSAL-D reports the highest AUBC on MNIST (0.9820), SVHN (0.8657), and TinyImageNet (0.2717), beating the clustering-based combined strategies BADGE and DBAL on three of the five datasets.
  • Adding the fixed prototype head to least confidence, margin, and entropy sampling improves or matches their AUBC on most settings in Table 3, so the diversity and uncertainty gains are not tied to the specific MSAL score.
  • Because the diversity step uses fixed prototypes instead of clustering, runtime stays only 6–10% above entropy sampling.
  • Pseudo-labeling with the prototype-based confidence (CEAL-MSAL) surpasses the entropy-based CEAL on all five datasets in AUBC and in final accuracy.
  • On long-tailed CIFAR-10 with imbalance factor 0.01, MSAL achieves the best final accuracy among compared uncertainty methods, though on long-tailed CIFAR-100 with imbalance factor 0.1 it trails margin sampling.

Reading between the lines

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

  • Beyond the paper: the diversity step's equal per-prototype allocation effectively enforces class balance in the query batch, so on highly imbalanced pools the number of rare-class samples that survive the uncertainty pre-filter will determine whether the method oversamples or starves those classes; this could be tested by varying $\beta$ per class.
  • Beyond the paper: the t-SNE stability across rounds suggests the fixed-prototype head may also preserve geometry in continual or semi-supervised learning, where the embedding space must not drift between data arrivals; an experiment would be to swap the uncertainty score for a small supervised signal and watch whether clusters stay fixed.
  • Beyond the paper: since the prototypes only need a fixed set of landmarks, the same uncertainty-plus-diversity recipe could be applied to pixel-level queries in segmentation or to object proposals in detection, which the paper lists as future work but does not test.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper proposes Maximally Separated Active Learning (MSAL), an uncertainty metric based on cosine similarity to fixed equiangular hyperspherical class prototypes, and MSAL-D, which adds diversity by pre-filtering the βb most uncertain samples and then selecting b/C samples closest to each prototype. The fixed prototypes come from the authors' prior maximum-separation work [33]. Experiments on MNIST, SVHN, CIFAR-10/100, and TinyImageNet compare against uncertainty-, diversity-, and combined-strategy baselines, with additional results on pseudo-labeling and long-tailed active learning. The central claims are that MSAL-D outperforms existing active learning methods across five benchmark datasets and that adding maximum separation boosts existing uncertainty strategies in both AUBC and final accuracy.

Significance. If the claims hold, the paper offers a simple and computationally cheap way to combine uncertainty and diversity without clustering, and it demonstrates that a fixed prototype head can stabilize representations throughout active learning. The work builds on a publicly available codebase and benchmarks against a standard suite of baselines, including BADGE, DBAL, and DeepAL+ implementations, which makes the empirical comparison reproducible in principle. The idea of using the same geometric structure both for uncertainty scoring and diversity allocation is elegant and potentially useful. However, the main significance is empirical, and the evidence as currently presented is mixed: several table entries contradict the stated conclusions, and the diversity component is not stress-tested under the class-imbalance setting where it is most fragile. The contribution is therefore promising but needs a more careful and honest empirical analysis before its claims can be accepted.

major comments (4)
  1. [Section 4.2, text after Table 3] The claim that 'making the active learning approach maximally separated boosts the performance across the datasets in both AUBC and F-acc metrics' is directly contradicted by Table 3. For example, LeastConf(MSAL) on SVHN has AUBC 0.8623 versus 0.8653 for the non-maximally-separated LeastConf baseline, and on CIFAR-100 the AUBC is 0.4743 versus 0.4750. Several MSAL-D rows also decrease relative to their non-MSAL counterparts (e.g., Entropy(MSAL-D) on CIFAR-10 AUBC 0.8103 versus 0.8120). Additionally, the baseline values differ between Table 2 and Table 3: Margin on CIFAR-10 is reported as 0.8147 in Table 2 but 0.8133 in Table 3. The authors should correct the numbers, report a single consistent baseline set, and temper the claim to state where max separation helps and where it does not.
  2. [Section 3.2 and Algorithm 1] The equal-allocation diversity step ('for c = 0 to C-1: find b/C closest samples') is the load-bearing component of MSAL-D, yet it is never evaluated under class imbalance. Table 5 reports only MSAL, not MSAL-D, so we do not know how the rule behaves when many classes have fewer than b/C candidates in the pre-filtered βb uncertain set. This is a concrete failure mode: with imbalance factor 0.1 or 0.01, rare-class prototypes may have very few or no nearby samples in the uncertain pool, making the per-prototype budget infeasible. The paper should either specify a fallback for such cases or include MSAL-D results on the long-tailed benchmarks, and discuss whether the rule remains representative under skew.
  3. [Table 2, CIFAR-10 row and abstract] The abstract claims outperforming 'various active learning methods across five benchmark datasets,' but on CIFAR-10 MSAL-D has AUBC 0.8107, which is below MSAL alone (0.8127) and below the Margin baseline (0.8147). Thus the diversity component actively hurts on one of the five datasets, and the overall claim is not supported. The authors should either restrict the claim to the datasets and metrics where MSAL-D is actually best, or analyze why the diversity step fails on CIFAR-10 (e.g., prototype-feature calibration, β choice, or per-class budget feasibility).
  4. [Section 4.2 and Table 2/Table 3] All empirical results are reported as point estimates from three runs, with no standard deviations, confidence intervals, or significance tests. The differences that support the central claims are often small on the AUBC scale (e.g., 0.9820 vs 0.9787 on MNIST, 0.4780 vs 0.4777 on CIFAR-100), so without variance information a reader cannot judge whether the improvements are meaningful or noise. For a paper whose main contribution is empirical, the authors should report run-level variation (at minimum variance or error bars) and preferably a paired test across the common random initializations.
minor comments (5)
  1. [Throughout] There are several typos and grammatical errors, including 'from from' (Section 3.2), 'stragegy' (Section 2.1), 'evalution' (Section 4), 'unlabled' (Section 2.1), 'ubiqutous' (Section 2.2), and 'the the' (Section 5). A careful proofreading pass is needed.
  2. [Figure 4 caption] The caption reads 'pre-filter factor on MNIST and SVHN datasets,' but the right panel is labeled 'pre-filter factor on CIFAR10.' The caption should match the figure contents.
  3. [Algorithm 1] The inner loop uses 'arg max_c α_MSAL' but the surrounding text and Figure 2 refer to selecting samples 'closest to each cluster center.' Since α_MSAL is a similarity score, 'arg max' and 'closest' are consistent, but this should be stated explicitly to avoid confusion; also 'for each class C' mixes the loop variable and the total number of classes.
  4. [Table 4] There is an extra space in some entries, e.g., '0 .9889' and '0 .9142,' which should be removed.
  5. [Section 4.4] The radius ρ is chosen as 0.1 for MNIST and 1 for other datasets 'using the heuristics as defined in [33],' but the paper does not report any sensitivity analysis for ρ. Since ρ scales the logits and directly affects the uncertainty scores, a brief ablation or justification would strengthen the claim that the method is not overly sensitive to this inherited hyperparameter.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the fixed prototype matrix is a parameter-free prior construction, the uncertainty and diversity steps are explicit heuristics, and all headline claims are benchmarked on held-out accuracy against external baselines.

full rationale

The paper's only self-citation is the fixed equiangular prototype matrix P_c and the radius heuristic taken from Kasarla et al. (NeurIPS 2022). That construction is a closed-form, parameter-free matrix given in Eq. (1); it is used as an architectural component and is not derived from, nor fitted to, the active-learning results. The proposed uncertainty score (Eq. 2) is the cosine similarity to these fixed prototypes, which is effectively the model's own confidence under the fixed classifier; this is a naming and architectural choice, not a case of predicting a fitted quantity. The diversity step in Algorithm 1 explicitly follows the prefilter-plus-cluster template of Zhdanov et al. with prototypes as fixed cluster centers; the equal per-prototype allocation is stated as a heuristic, and its effect is reported on held-out test accuracy rather than claimed to be forced by the prototype geometry. Tables 2 and 3 compare against independent baselines such as Margin, Entropy, BADGE, DBAL, and KCenter on AUBC and final accuracy, so the reported numbers are not obtained by plugging the same data back into the selection rule. The long-tailed experiments in Table 5 report only MSAL, and the paper itself concedes underperformance on CIFAR-100 with imbalance factor 0.1; that is a correctness or robustness caveat, not circularity. No uniqueness theorem is imported to forbid alternative selection rules, and no equation reduces a claimed prediction to its own input. The fixed-prototype construction could in principle be wrong or inapplicable, but the paper's central empirical claims remain independently testable against external benchmarks, so the derivation chain is not circular.

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

The central claim rests on a small number of tuned hyperparameters (beta, rho, b) and on three domain assumptions about the fixed prototypes: that they are maximally separated, that the network aligns to them, and that prototype-proximity is a sufficient diversity cover. No new entities are introduced by this paper; the prototypes are inherited from prior work.

free parameters (3)
  • pre-filter factor beta = 5 (MNIST, SVHN), 10 (CIFAR-10/100, TinyImageNet)
    Chosen by grid search on the accuracy-vs-budget curve for MNIST ({5,10,20,25,30}) and CIFAR-10 ({5,10,15}); it controls how many uncertain samples enter the diversity step.
  • hypersphere radius rho = 0.1 (MNIST), 1 (others)
    Taken as a heuristic from the authors' prior work [33]; scales logits before softmax and affects the uncertainty scores.
  • query batch size b = 250 (MNIST/SVHN), 500 (CIFAR/TinyImageNet)
    Set per dataset in Table 1; ablation on CIFAR-10 selected 500 by highest final accuracy.
assumptions (3)
  • domain assumption The fixed equiangular prototype matrix P_c provides maximally separated, uniformly distributed class directions in R^(C-1).
    Inherited from Kasarla et al. [33] without re-derivation in this paper; all uncertainty and diversity scores depend on this property. Invoked in Section 3.1.
  • domain assumption Features learned with the fixed head align with the nearest prototype, so cosine similarity to the closest prototype is a valid uncertainty measure.
    The whole MSAL uncertainty ranking in Section 3.2 assumes the trained model clusters by class around the fixed prototypes; if the head does not align, Eq. 2 is not a meaningful uncertainty score.
  • ad hoc to paper Selecting b/C samples closest to each prototype from the beta-b most uncertain samples yields a diverse and representative batch.
    This heuristic in Algorithm 1 replaces clustering; the equal per-class allocation is assumed valid even for imbalanced distributions, which Table 5 shows is questionable.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Maximally Separated Active Learning." pith.science (2026). https://pith.science/paper/PZ6CHPLU

@misc{pith2026241117444,
  author       = {Pith},
  title        = {Pith review of: Maximally Separated Active Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PZ6CHPLU}},
  note         = {Machine review of arXiv:2411.17444}
}
read the original abstract

Active Learning aims to optimize performance while minimizing annotation costs by selecting the most informative samples from an unlabelled pool. Traditional uncertainty sampling often leads to sampling bias by choosing similar uncertain samples. We propose an active learning method that utilizes fixed equiangular hyperspherical points as class prototypes, ensuring consistent inter-class separation and robust feature representations. Our approach introduces Maximally Separated Active Learning (MSAL) for uncertainty sampling and a combined strategy (MSAL-D) for incorporating diversity. This method eliminates the need for costly clustering steps, while maintaining diversity through hyperspherical uniformity. We demonstrate strong performance over existing active learning techniques across five benchmark datasets, highlighting the method's effectiveness and integration ease. The code is available on GitHub.

Figures

Figures reproduced from arXiv: 2411.17444 by the authors.

Figure 1
Figure 1. Overall accuracy vs. budget curves on MNIST and TinyImageNet datasets. sampling for (MSAL-D). Across all settings, as reported in [PITH_FULL_IMAGE:figures/full_fig_p009_1.png] view at source ↗
Figure 2
Figure 2. A visualization of embeddings and selected samples for labeling. [PITH_FULL_IMAGE:figures/full_fig_p011_2.png] view at source ↗
Figure 3
Figure 3. Percentage of misclassified unlabeled samples [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Overall accuracy vs. budget curves for β on MNIST and SVHN datasets. CIFAR-10 shown in [PITH_FULL_IMAGE:figures/full_fig_p013_4.png]
Figure 5
Figure 5. Figure 5: Ablation of AUBC and Accuracy(F-acc) vs query batch size per active learning round. We choose the batchsize with highest F-acc. 5 Conclusion This paper aims to integrate hyperspherical uniformity through maximum sepa￾ration into active learning for image classification…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

96 extracted references · 64 canonical work pages

  1. [33]

    In: Advances in Neural Information Processing Systems (2022)

    Kasarla, T., Burghouts, G.J., van Spengler, M., van der Pol, E., Cucchiara, R., Mettes, P.: Maximum class separation as inductive bias in one matrix. In: Advances in Neural Information Processing Systems (2022)

    work page 2022

  2. [1]

    Adiloglu, K., Annies, R., Henrich, F.F., Paus, A., Obermayer, K.: Geometrical approaches to active learning. In: Autonomous Systems–Self-Organization, Man- agement, and Control: Proceedings of the 8th International Workshop held at Shanghai Jiao Tong University, Shanghai, China, October 6–7, 2008. pp. 11–19. Springer (2008)

    2008

  3. [2]

    arXiv preprint arXiv:1906.03671 (2019)

    Ash, J.T., Zhang, C., Krishnamurthy, A., Langford, J., Agarwal, A.: Deep batch active learning by diverse, uncertain gradient lower bounds. arXiv preprint arXiv:1906.03671 (2019)

    arXiv 2019

  4. [3]

    In: International Con- ference on Learning Representations (2020)

    Ash, J.T., Zhang, C., Krishnamurthy, A., Langford, J., Agarwal, A.: Deep batch active learning by diverse, uncertain gradient lower bounds. In: International Con- ference on Learning Representations (2020)

    2020

  5. [4]

    HYPO: Hyperspherical Out-of-Distribution Generalization

    Bai, H., Ming, Y., Katz-Samuels, J., Li, Y.: Hypo: Hyperspherical out-of- distribution generalization. arXiv preprint arXiv:2402.07785 (2024)

    work page Pith review arXiv 2024

  6. [5]

    In: Learning Theory: 20th Annual Conference on Learning Theory, COLT 2007, San Diego, CA, USA; June 13-15, 2007

    Balcan, M.F., Broder, A., Zhang, T.: Margin based active learning. In: Learning Theory: 20th Annual Conference on Learning Theory, COLT 2007, San Diego, CA, USA; June 13-15, 2007. Proceedings 20. pp. 35–50. Springer (2007)

    2007

  7. [6]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

    Beluch, W.H., Genewein, T., Nürnberger, A., Köhler, J.M.: The power of ensembles for active learning in image classification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). pp. 9368–9377 (2018)

    2018

  8. [7]

    In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision

    Bengar, J.Z., van de Weijer, J., Fuentes, L.L., Raducanu, B.: Class-balanced ac- tive learning for image classification. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. pp. 1536–1545 (2022) Maximally Separated Active Learning 15

    2022

Show all 96 references
  1. [8]

    In: International Conference on Machine Learning (2003)

    Brinker, K.: Incorporating diversity in active learning with support vector ma- chines. In: International Conference on Machine Learning (2003)

    2003

  2. [9]

    IEEE Transactions on Pattern Analysis and Machine Intelligence 45(11), 13422–13437 (2023)

    Cao, X., Liu, W., Tsang, I.W.: Data-efficient learning via minimizing hyperspher- ical energy. IEEE Transactions on Pattern Analysis and Machine Intelligence 45(11), 13422–13437 (2023)

    2023

  3. [10]

    In: Asian Conference on Machine Learning

    Cevikalp, H., Yavuz, H.S., Saribas, H.: Deep uniformly distributed centers on a hypersphere for open set recognition. In: Asian Conference on Machine Learning. pp. 217–230. PMLR (2024)

    2024

  4. [11]

    In: Beygelzimer, A., Dauphin, Y., Liang, P., Vaughan, J.W

    Citovsky, G., DeSalvo, G., Gentile, C., Karydas, L., Rajagopalan, A., Ros- tamizadeh, A., Kumar, S.: Batch active learning at scale. In: Beygelzimer, A., Dauphin, Y., Liang, P., Vaughan, J.W. (eds.) Advances in Neural Information Processing Systems (2021)

    2021

  5. [12]

    Cohn,H.,Kumar,A.:Universallyoptimaldistributionofpointsonspheres.Journal of the American Mathematical Society20(1), 99–148 (2007)

    2007

  6. [13]

    arXiv preprint arXiv:2205.06926 (2022)

    Cosentino,R.,Sengupta,A.,Avestimehr,S.,Soltanolkotabi,M.,Ortega,A.,Willke, T., Tepper, M.: Toward a geometrical understanding of self-supervised contrastive learning. arXiv preprint arXiv:2205.06926 (2022)

    2022 arXiv

  7. [14]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

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

    2019

  8. [15]

    Theoretical Computer Science412(19), 1767–1781 (2011), algorithmic Learning Theory (ALT 2009)

    Dasgupta, S.: Two faces of active learning. Theoretical Computer Science412(19), 1767–1781 (2011), algorithmic Learning Theory (ALT 2009)

    2011

  9. [16]

    IEEE signal processing magazine29(6), 141–142 (2012)

    Deng, L.: The mnist database of handwritten digit images for machine learning research [best of the web]. IEEE signal processing magazine29(6), 141–142 (2012)

    2012

  10. [17]

    arXiv preprint arXiv:1802.09841 (2018)

    Ducoffe, M., Precioso, F.: Adversarial active learning for deep networks: a margin based approach. arXiv preprint arXiv:1802.09841 (2018)

    2018 arXiv

  11. [18]

    Image and Vision Computing124, 104494 (2022)

    Durrant, A., Leontidis, G.: Hyperspherically regularized networks for self- supervision. Image and Vision Computing124, 104494 (2022)

    2022

  12. [19]

    In: IEEE Intelligent Vehicles Symposium (IV)

    Feng, D., Wei, X., Rosenbaum, L., Maki, A., Dietmayer, K.: Deep active learning for efficient training of a lidar 3d object detector. In: IEEE Intelligent Vehicles Symposium (IV). pp. 667–674. IEEE (2019)

    2019

  13. [20]

    Freeman, L.C.: Elementary applied statistics: for students in behavioral science (1965)

    1965

  14. [21]

    In: Proceedings on European Conference on Computer Vision

    Freytag, A., Rodner, E., Denzler, J.: Selecting influential examples: Active learning with expected model output changes. In: Proceedings on European Conference on Computer Vision. pp. 562–577. Springer (2014)

    2014

  15. [22]

    Knowl- edge and information systems35, 249–283 (2013)

    Fu, Y., Zhu, X., Li, B.: A survey on instance selection for active learning. Knowl- edge and information systems35, 249–283 (2013)

    2013

  16. [23]

    In: International Conference on Machine Learning

    Gal, Y., Islam, R., Ghahramani, Z.: Deep bayesian active learning with image data. In: International Conference on Machine Learning. pp. 1183–1192 (2017)

    2017

  17. [24]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    Gudovskiy, D., Hodgkinson, A., Yamaguchi, T., Tsukizawa, S.: Deep active learn- ing for biased datasets via fisher kernel self-supervision. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). pp. 9041–9049 (2020)

    2020

  18. [25]

    In: IEEE Intelligent Vehicles Symposium (IV)

    Haussmann, E., Fenzi, M., Chitta, K., Ivanecky, J., Xu, H., Roy, D., Mittel, A., Koumchatzky, N., Farabet, C., Alvarez, J.M.: Scalable active learning for object detection. In: IEEE Intelligent Vehicles Symposium (IV). pp. 1430–1435. IEEE (2020)

    2020

  19. [26]

    He,K.,Zhang,X.,Ren,S.,Sun,J.:Deepresiduallearningforimagerecognition.In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). pp. 770–778 (2016) 16 Kasarla et al

    2016

  20. [27]

    Journal of Machine Learning Research9(1) (2008)

    Henrich, F.F., Obermayer, K.: Active learning by spherical subdivision. Journal of Machine Learning Research9(1) (2008)

    2008

  21. [28]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Hua, T., Wang, W., Xue, Z., Ren, S., Wang, Y., Zhao, H.: On feature decorrela- tion in self-supervised learning. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 9598–9608 (2021)

    2021

  22. [29]

    In: International Conference on Discovery Science

    Ienco, D., Bifet, A., Žliobait˙ e, I., Pfahringer, B.: Clustering based active learning for evolving data streams. In: International Conference on Discovery Science. pp. 79–93. Springer (2013)

    2013

  23. [30]

    arXiv preprint arXiv:2402.14957 (2024)

    Jha, A., Blaschko, M.B., Asano, Y.M., Tuytelaars, T.: The common stabil- ity mechanism behind most self-supervised learning approaches. arXiv preprint arXiv:2402.14957 (2024)

    2024 arXiv

  24. [31]

    IEEE Transactions on Visualization and Computer Graph- ics 28(1), 791–801 (2021)

    Jia, S., Li, Z., Chen, N., Zhang, J.: Towards visual explainable active learning for zero-shot classification. IEEE Transactions on Visualization and Computer Graph- ics 28(1), 791–801 (2021)

    2021

  25. [32]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016)

    Kampffmeyer, M., Salberg, A.B., Jenssen, R.: Semantic segmentation of small ob- jects and modeling of uncertainty in urban remote sensing images using deep con- volutional neural networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016)

    2016

  26. [34]

    In: Proceedings of the IEEE Winter Conference on Applications of Computer Vision (WACV)

    Kasarla, T., Nagendar, G., Hegde, G.M., Balasubramanian, V., Jawahar, C.: Region-based active learning for efficient labeling in semantic segmentation. In: Proceedings of the IEEE Winter Conference on Applications of Computer Vision (WACV). pp. 1109–1117. IEEE (2019)

    2019

  27. [35]

    arXiv preprint arXiv:2011.13570 (2020)

    Kim, Y.: Deep active learning for sequence labeling based on diversity and uncer- tainty in gradient. arXiv preprint arXiv:2011.13570 (2020)

    2020 arXiv

  28. [36]

    arXiv preprint arXiv:1412.6980 (2014)

    Kingma, D.P., Ba, J.: Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)

    2014 arXiv

  29. [37]

    Advances in Neural Information Pro- cessing Systems32 (2019)

    Kirsch, A., Van Amersfoort, J., Gal, Y.: Batchbald: Efficient and diverse batch acquisition for deep bayesian active learning. Advances in Neural Information Pro- cessing Systems32 (2019)

    2019

  30. [38]

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

    2009

  31. [39]

    CS 231N7(7), 3 (2015)

    Le, Y., Yang, X.: Tiny imagenet visual recognition challenge. CS 231N7(7), 3 (2015)

    2015

  32. [40]

    In: Acm Sigir Forum

    Lewis, D.D.: A sequential algorithm for training text classifiers: Corrigendum and additional data. In: Acm Sigir Forum. vol. 29, pp. 13–19. ACM New York, NY, USA (1995)

    1995

  33. [41]

    ACM Journal of Data and Information Quality (2024)

    Li, N., Qi, Y., Li, C., Zhao, Z.: Active learning for data quality control: A survey. ACM Journal of Data and Information Quality (2024)

    2024

  34. [42]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2020)

    Lin, R., Liu, W., Liu, Z., Feng, C., Yu, Z., Rehg, J.M., Xiong, L., Song, L.: Reg- ularizing neural networks via minimizing hyperspherical energy. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2020)

    2020

  35. [43]

    arXiv preprint arXiv:1711.01732 (2017)

    Lin, X., Parikh, D.: Active learning for visual question answering: An empirical study. arXiv preprint arXiv:1711.01732 (2017)

    2017 arXiv

  36. [44]

    NeurIPS (2018) Maximally Separated Active Learning 17

    Liu, W., Lin, R., Liu, Z., Liu, L., Yu, Z., Dai, B., Song, L.: Learning towards minimum hyperspherical energy. NeurIPS (2018) Maximally Separated Active Learning 17

    2018

  37. [45]

    In: International Conference on Artificial Intelligence and Statistics (AISTATS) (2021)

    Liu, W., Lin, R., Liu, Z., Xiong, L., Schölkopf, B., Weller, A.: Learning with hy- perspherical uniformity. In: International Conference on Artificial Intelligence and Statistics (AISTATS) (2021)

    2021

  38. [46]

    arXiv preprint arXiv:1612.02295 (2016)

    Liu, W., Wen, Y., Yu, Z., Yang, M.: Large-margin softmax loss for convolutional neural networks. arXiv preprint arXiv:1612.02295 (2016)

    2016 arXiv

  39. [47]

    In: Advances in Neural Information Processing Systems (2017)

    Liu, W., Zhang, Y.M., Li, X., Yu, Z., Dai, B., Zhao, T., Song, L.: Deep hyper- spherical learning. In: Advances in Neural Information Processing Systems (2017)

    2017

  40. [48]

    In: International Conference on Learning Representations (2024)

    Lu, H., Gong, D., Wang, S., Xue, J., Yao, L., Moore, K.: Learning with mix- ture of prototypes for out-of-distribution detection. In: International Conference on Learning Representations (2024)

    2024

  41. [49]

    In: International Conference on Learning Representations (2021)

    Mahmood, R., Fidler, S., Law, M.T.: Low-budget active learning via wasser- stein distance: An integer programming approach. In: International Conference on Learning Representations (2021)

    2021

  42. [50]

    NeurIPS (2019)

    Mettes, P., van der Pol, E., Snoek, C.G.M.: Hyperspherical prototype networks. NeurIPS (2019)

    2019

  43. [51]

    Ming, Y., Sun, Y., Dia, O., Li, Y.: How to exploit hyperspherical embeddings for out-of-distribution detection? In: International Conference on Learning Represen- tations (2023)

    2023

  44. [52]

    ACM Comput

    Mohseni, S., Wang, H., Xiao, C., Yu, Z., Wang, Z., Yadawa, J.: Taxonomy of machine learning safety: A survey and primer. ACM Comput. Surv.55(8) (dec 2022)

    2022

  45. [53]

    Advances in Neural Information Processing Systems25 (2012)

    Mroueh, Y., Poggio, T., Rosasco, L., Slotine, J.J.: Multiclass learning with simplex coding. Advances in Neural Information Processing Systems25 (2012)

    2012

  46. [54]

    In: NIPS workshop on deep learning and unsupervised feature learning (2011)

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

    2011

  47. [55]

    Nguyen, V.L., Shaker, M.H., Hüllermeier, E.: How to measure uncertainty in uncer- taintysamplingforactivelearning.In:InternationConferenceonMachineLearning (2022)

    2022

  48. [56]

    IEEE Transactions on Neural Networks and Learning Systems pp

    Pernici, F., Bruni, M., Baecchi, C., Bimbo, A.D.: Regular polytope networks. IEEE Transactions on Neural Networks and Learning Systems pp. 1–15 (2021)

    2021

  49. [57]

    International Conference on Machine Learning2(441- 448), 4 (2001)

    Roy, N., McCallum, A.: Toward optimal active learning through monte carlo esti- mation of error reduction. International Conference on Machine Learning2(441- 448), 4 (2001)

    2001

  50. [58]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    Saadabadi, M.S.E., Dabouei, A., Malakshan, S.R., Nasrabadi, N.M.: Hyperspher- ical classification with dynamic label-to-prototype assignment. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). pp. 17333–17342 (2024)

    2024

  51. [59]

    arXiv preprint arXiv:2008.07267 (2020)

    Schröder, C., Niekler, A.: A survey of active learning for text classification using deep neural networks. arXiv preprint arXiv:2008.07267 (2020)

    2020 arXiv

  52. [60]

    In: International Conference on Learning Representations (2018)

    Sener, O., Savarese, S.: Active learning for convolutional neural networks: A core- set approach. In: International Conference on Learning Representations (2018)

    2018

  53. [61]

    Settles, B.: Active learning literature survey (2009)

    2009

  54. [62]

    In: Active learning and experimental design workshop (2010)

    Settles, B.: From theories to queries: Active learning in practice. In: Active learning and experimental design workshop (2010)

    2010

  55. [63]

    In: Proceedings of the Fifth Annual Workshop on Computational Learning Theory

    Seung, H.S., Opper, M., Sompolinsky, H.: Query by committee. In: Proceedings of the Fifth Annual Workshop on Computational Learning Theory. pp. 287–294 (1992)

    1992

  56. [64]

    ACM SIGMOBILE Mo- bile Computing and Communications Review5(1), 3–55 (2001) 18 Kasarla et al

    Shannon, C.E.: A mathematical theory of communication. ACM SIGMOBILE Mo- bile Computing and Communications Review5(1), 3–55 (2001) 18 Kasarla et al

    2001

  57. [65]

    Data Mining and Knowledge Discovery31, 164–202 (2017)

    Sharma, M., Bilgic, M.: Evidence-based uncertainty sampling for active learning. Data Mining and Knowledge Discovery31, 164–202 (2017)

    2017

  58. [66]

    In: International Conference on Artificial Intelligence and Statistics

    Shui, C., Zhou, F., Gagné, C., Wang, B.: Deep active learning: Unified and prin- cipled method for query and training. In: International Conference on Artificial Intelligence and Statistics. pp. 1308–1318. PMLR (2020)

    2020

  59. [67]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Sinha, S., Ebrahimi, S., Darrell, T.: Variational adversarial active learning. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 5972–5981 (2019)

    2019

  60. [68]

    arXiv preprint arXiv:1912.00594 (2019)

    Song, S., Berthelot, D., Rostamizadeh, A.: Combining mixmatch and active learn- ing for better accuracy with fewer labels. arXiv preprint arXiv:1912.00594 (2019)

    2019 arXiv

  61. [69]

    Mathematics11(4), 820 (2023)

    Tharwat, A., Schenck, W.: A survey on active learning: State-of-the-art, practical challenges and research directions. Mathematics11(4), 820 (2023)

    2023

  62. [70]

    Thomson, J.J.: Xxiv. on the structure of the atom: an investigation of the stability and periods of oscillation of a number of corpuscles arranged at equal intervals around the circumference of a circle; with application of the results to the theory of atomic structure. The Lo...

    1904

  63. [71]

    In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

    Vezhnevets, A., Buhmann, J.M., Ferrari, V.: Active learning for semantic segmen- tation with expected change. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR). pp. 3162–3169. IEEE (2012)

    2012

  64. [72]

    In: Interna- tional Joint Conference on Neural Networks (IJCNN)

    Wang, D., Shang, Y.: A new active labeling method for deep learning. In: Interna- tional Joint Conference on Neural Networks (IJCNN). pp. 112–119. IEEE (2014)

    2014

  65. [73]

    Medical Image Analysis p

    Wang, H., Jin, Q., Li, S., Liu, S., Wang, M., Song, Z.: A comprehensive survey on deep active learning in medical image analysis. Medical Image Analysis p. 103201 (2024)

    2024

  66. [74]

    IEEE Transactions on Circuits and Systems for Video Technology27(12), 2591–2600 (2016)

    Wang, K., Zhang, D., Li, Y., Zhang, R., Lin, L.: Cost-effective active learning for deep image classification. IEEE Transactions on Circuits and Systems for Video Technology27(12), 2591–2600 (2016)

    2016

  67. [75]

    In: International Conference on Machine Learning

    Wang, T., Isola, P.: Understanding contrastive representation learning through alignment and uniformity on the hypersphere. In: International Conference on Machine Learning. pp. 9929–9939 (2020)

    2020

  68. [76]

    Advances in Neural Infor- mation Processing Systems (2020)

    Wang, Z., Xiang, C., Zou, W., Xu, C.: Mma regularization: Decorrelating weights of neural networks by maximizing the minimal angles. Advances in Neural Infor- mation Processing Systems (2020)

    2020

  69. [77]

    In: International Conference on Machine Learning (2015)

    Wei, K., Iyer, R., Bilmes, J.: Submodularity in data subset selection and active learning. In: International Conference on Machine Learning (2015)

    2015

  70. [78]

    Woo, J.O.: Active learning in bayesian neural networks with balanced entropy learningprinciple.In:InternationalConferenceonLearningRepresentations(2023)

    2023

  71. [79]

    In: 2006 IEEE International Conference on Multimedia and Expo

    Wu, Y., Kozintsev, I., Bouguet, J.Y., Dulong, C.: Sampling strategies for active learning in personal photo retrieval. In: 2006 IEEE International Conference on Multimedia and Expo. pp. 529–532. IEEE (2006)

    2006

  72. [80]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    Xie, B., Yuan, L., Li, S., Liu, C.H., Cheng, X.: Towards fewer annotations: Active learning via region impurity and prediction uncertainty for domain adaptive se- mantic segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)....

    2022

  73. [81]

    In: Medical Image Computing and Computer Assisted Intervention- (MICCAI)

    Yang, L., Zhang, Y., Chen, J., Zhang, S., Chen, D.Z.: Suggestive annotation: A deep active learning framework for biomedical image segmentation. In: Medical Image Computing and Computer Assisted Intervention- (MICCAI). pp. 399–407. Springer (2017)

    2017

  74. [82]

    IEEE Access9, 51452–51463 (2021) Maximally Separated Active Learning 19

    Yang, Y., Yin, X., Zhao, Y., Lei, J., Li, W., Shu, Z.: Batch mode active learning based on multi-set clustering. IEEE Access9, 51452–51463 (2021) Maximally Separated Active Learning 19

    2021

  75. [83]

    (eds.) Advances in Neural Information Processing Systems (2022)

    Yang, Y., Chen, S., Li, X., Xie, L., Lin, Z., Tao, D.: Inducing neural collapse in imbalanced learning: Do we really need a learnable classifier at the end of deep neural network? In: Oh, A.H., Agarwal, A., Belgrave, D., Cho, K. (eds.) Advances in Neural Information Processing...

    2022

  76. [84]

    Yehuda, O., Dekel, A., Hacohen, G., Weinshall, D.: Active learning through a coveringlens.AdvancesinNeuralInformationProcessingSystems 35,22354–22367 (2022)

    2022

  77. [85]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

    Yoo, D., Kweon, I.S.: Learning loss for active learning. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). pp. 93–102 (2019)

    2019

  78. [86]

    arXiv preprint arXiv:2107.01622 (2021)

    Zhan, X., Li, Q., Chan, A.B.: Multiple-criteria based active learning with fixed-size determinantal point processes. arXiv preprint arXiv:2107.01622 (2021)

    2021 arXiv

  79. [87]

    In: IJCAI

    Zhan, X., Liu, H., Li, Q., Chan, A.B.: A comparative survey: Benchmarking for pool-based active learning. In: IJCAI. pp. 4679–4686 (2021)

    2021

  80. [88]

    arXiv preprint arXiv:2203.13450 (2022)

    Zhan, X., Wang, Q., Huang, K.h., Xiong, H., Dou, D., Chan, A.B.: A comparative survey of deep active learning. arXiv preprint arXiv:2203.13450 (2022)

    2022 arXiv

  81. [89]

    In: Proceedings of the Thirty-First AAAI Conference on Artificial Intel- ligence

    Zhang, Y., Lease, M., Wallace, B.C.: Active discriminative text representation learning. In: Proceedings of the Thirty-First AAAI Conference on Artificial Intel- ligence. pp. 3386–3392 (2017)

    2017

  82. [90]

    Advances in Neural In- formation Processing Systems34, 9734–9746 (2021)

    Zhao, G., Dougherty, E., Yoon, B.J., Alexander, F., Qian, X.: Efficient active learn- ing for gaussian process classification by error reduction. Advances in Neural In- formation Processing Systems34, 9734–9746 (2021)

    2021

  83. [91]

    In: International Conference on Learning Representations (2021)

    Zhao, G., Dougherty, E., Yoon, B.J., Alexander, F., Qian, X.: Uncertainty-aware active learning for optimal bayesian classifier. In: International Conference on Learning Representations (2021)

    2021

  84. [92]

    arXiv preprint arXiv:1901.05954 (2019)

    Zhdanov, F.: Diverse mini-batch active learning. arXiv preprint arXiv:1901.05954 (2019)

    2019 arXiv

  85. [93]

    arXiv preprint arXiv:2208.00789 (2022)

    Zheng, L., Puy, G., Riccietti, E., Pérez, P., Gribonval, R.: Self-supervised learning with rotation-invariant kernels. arXiv preprint arXiv:2208.00789 (2022)

    2022 arXiv

  86. [94]

    In: International Conference on Learning Representations (2022)

    Zhou, X., Liu, X., Zhai, D., Jiang, J., Gao, X., Ji, X.: Learning towards the largest margins. In: International Conference on Learning Representations (2022)

    2022

  87. [95]

    arXiv preprint arXiv:1702.07956 (2017)

    Zhu, J.J., Bento, J.: Generative adversarial active learning. arXiv preprint arXiv:1702.07956 (2017)

    2017 arXiv

  88. [96]

    Advances in Neural Information Processing Systems34, 29820–29834 (2021)

    Zhu, Z., Ding, T., Zhou, J., Li, X., You, C., Sulam, J., Qu, Q.: A geometric analysis of neural collapse with unconstrained features. Advances in Neural Information Processing Systems34, 29820–29834 (2021)

    2021

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.