Pith. sign in

REVIEW 4 major objections 5 minor 41 references

O-MedAL: Online Active Deep Learning for Medical Image Analysis

T0 review · 4 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read An online active-learning method reaches its baseline accuracy with about a quarter of the labels by querying images farthest from labeled examples in feature space.

desk verdict The paper's central speedup theorem is false, so the headline claim is unsupported, though the underlying active-learning combination might be worth a careful second look. read the letter →

arxiv 1908.10508 v2 pith:M3BLYA7Q submitted 2019-08-28 cs.LG cs.CVeess.IVstat.ML

Asim Smailagic , Pedro Costa , Alex Gaudio , Kartik Khandelwal , Mostafa Mirshekari , Jonathon Fagert , Devesh Walawalkar , Susu Xu
show 4 more authors
Adrian Galdran Pei Zhang Aurélio Campilho Hae Young Noh
This is my paper · ORCID
classification cs.LGcs.CVeess.IVstat.ML
keywords activelearningonlinedeepmedicalimageanalysisfeatureembeddingsuncertaintysamplingclassimbalancediabeticretinopathy
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

O-MedAL is an online active learning method for medical image classification. It claims that an unlabeled image is worth labeling when it is far in the network's learned feature space from everything already labeled, and that this criterion can be queried cheaply: selecting the example farthest from the centroid of the labeled examples preserves the ranking of the average-distance score. The paper reports that this lets a ResNet-18 reach its fully-supervised baseline accuracy with only about 25% of the dataset labeled, improve on that baseline by 6.30% when more labels are used, and cut backpropagated images by up to roughly 68%. The practical stake is that medical experts labeling images can interact with the model in near-real-time instead of waiting for full retraining between batches.

What carries the argument

The central object is the score $s(X) = \frac{1}{N}\sum_{i=1}^N d(f(X_i), f(X))$, the average Euclidean distance from an unlabeled image's feature embedding to all labeled embeddings. The linchpin is Theorem 1, which replaces the pairwise sums by a single distance to the centroid $Y=\frac{1}{N}\sum_i f(X_i)$: the argmax over candidates is claimed to be unchanged, turning an $O(MN)$ pairwise computation into $O(M)$ distances. A second mechanism is the online training schedule: the weights are never reset, and each iteration trains on the newly labeled batch plus a $p$-fraction of previously labeled examples, so the model improves incrementally rather than repeatedly relearning from scratch.

What would settle it

Compute the average Euclidean distance and the centroid distance for a one-dimensional example with labeled points $\{0,1,10\}$ and candidates $\{2,5\}$. The average-distance score ranks 5 above 2 ($14/3 > 11/3$), while the centroid-distance score ranks 2 above 5 ($5/3 > 4/3$). This is a direct counterexample to the claimed rank preservation for plain Euclidean distance; if the actual sampler uses plain Euclidean distance, the $O(M)$ shortcut can select a different image than the stated criterion.

Watch

Extended reading notes

Core claim

The paper claims that active learning for deep networks can be made both label-efficient and computationally cheap by combining predictive-entropy screening with a distance criterion in feature space. At each active-learning iteration, the top-$M$ unlabeled images by predictive entropy form a candidate set $C_M$; among those, O-MedAL selects the image maximizing the average Euclidean distance to all labeled examples in the embedding $f(\cdot)$. Theorem 1 states that this argmax can be evaluated as the distance to the centroid of the labeled embeddings, reducing the per-iteration cost from $O(MN)$ to $O(M)$ distance evaluations. The online component then fine-tunes the same network on the newly labeled images plus a random subset of previously labeled images, without resetting weights between iterations. On Messidor diabetic retinopathy data, O-MedAL with $p=0.875$ and early stopping reaches 92.02% test accuracy versus an 85.71% ResNet-18 baseline, reaches the baseline accuracy with 25.29% of labels, and the most computation-efficient configuration processes about 68% fewer backpropagated images than the baseline.

Load-bearing premise

The load-bearing premise is that distance in the learned feature space measures how much an unlabeled image will improve the classifier, and that ranking examples by average Euclidean distance to all labeled examples is the same as ranking them by distance to the centroid; the proof establishes this exactly for squared Euclidean distances, while the paper states the score with plain Euclidean distance.

Editorial extensions

If this is right

  • If the centroid-ranking claim holds, the sampling step scales to large unlabeled pools, and batch size can be reduced toward single-image queries without quadratic cost, enabling near-real-time human-in-the-loop annotation.
  • If O-MedAL's empirical results generalize, a medical imaging team can obtain a network's fully supervised accuracy after labeling a quarter of the data, with a human expert spending far less time on annotation.
  • The online fine-tuning schedule means the model is updated incrementally, so annotators do not wait between active-learning iterations; this is the practical condition for using active learning in a clinical workflow.
  • Consistent gains over uncertainty-only and random sampling on balanced, imbalanced, binary, and multi-class medical datasets suggest the method is not specialized to one disease or imaging modality.
  • Because selected subsets can outperform the fully supervised baseline, the method also points toward identifying training examples that are redundant or unhelpful, not merely toward spending fewer labels.

Reading between the lines

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

  • As an extension beyond the paper, applying the same farthest-from-centroid sampling to non-medical image datasets would test whether the method is a general active-learning recipe or specific to medical imaging.
  • The observation that subsets of labels can beat the fully supervised baseline suggests O-MedAL could be turned into a data-subset selection tool for deep learning generally, identifying redundant or harmful training examples.
  • Because the model is updated online without resets, the framework could also serve streaming or continuously arriving data, where new classes or distributions arrive over time; that setting is not the paper's focus.
  • Batch size can shrink toward single-image queries without quadratic cost, which in practice would let an annotator label one image, see the model update, and immediately receive the next query.
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 paper extends the authors' MedAL active learning framework to an online setting, O-MedAL, which keeps model weights between iterations, trains on newly labeled examples plus a random subset of previously labeled data, and selects examples with a distance-based sampling rule. The central claims are: (i) maximizing the average Euclidean distance from an unlabeled example to all labeled examples can be replaced, without changing the ranking, by maximizing the Euclidean distance to the centroid of the labeled examples, giving an O(M) sampler; (ii) online training increases accuracy over a ResNet18 baseline by 6.30% while using only 25.29% of the labeled data to reach baseline accuracy; and (iii) these gains generalize across binary/multi-class and balanced/imbalanced medical image tasks. The experiments are conducted on Messidor, BACH breast cancer, and ISIC skin lesion datasets.

Significance. If the central claims were valid, the linear-time exact reformulation of the sampling rule would be a practically useful contribution to active learning for deep medical image analysis, and the reported label-efficiency gains would be of interest to the clinical machine-learning community. Strengths include the use of three public medical datasets and the release of public code. However, the main theoretical equivalence is not proven as stated, and the headline results appear to be obtained under evaluation protocols that do not support them. The significance of the contributions is therefore not established by this manuscript.

major comments (4)
  1. [Proposed Method, Theorem 1 and Eqs. (5)-(9)] The proof of rank preservation is invalid. The text moves from the scalar property that squaring a nonnegative quantity preserves its order to the claim that the order of sums of square roots is preserved after squaring each term before summing; this is not true. A one-dimensional counterexample is labeled embeddings {0,10,10} with candidates x=6 and x=8. The average Euclidean distances are s(6)=14/3 and s(8)=4, so x=6 is the maximizer, while the distances to the centroid 20/3 are 2/3 and 4/3, so x=8 is the maximizer. The rankings are reversed. Consequently, the O(M) exact evaluation of the average-distance rule in Eq. (2) is not established, and the sampler described in Figure 1, which selects examples farthest from the centroid, is a different heuristic from the one in Eq. (2). The equivalence would hold for squared Euclidean distances, but the paper explicitly selects Euclidean distance.
  2. [O-MedAL Evaluation, Figure 5A] The hyperparameters p and the patience values appear to be selected on the same Messidor test set that produces the headline claims. The text states that the p=0.125 and p=0.875 models were further evaluated with patience values of 5, 10, and 20 'as these two values of p were best performing,' and no validation split for O-MedAL is described. Because the reported table gives maximum test accuracy, the headline 92.02%, 25.29%, and 6.30% figures are likely optimistic selections on the test set. This is a circular evaluation of the main accuracy claims.
  3. [O-MedAL Implementation Details, Messidor split] The paper states that for Messidor the 80/20 split is 're-computed each time a model is trained.' This means the test set is not fixed across the compared models or across iterations, so the test accuracies in Figures 4-5 and Table 5A are not measured on a common test set. The comparison between O-MedAL, MedAL, and the ResNet18 baseline is therefore not statistically clean, and the reported +6.30% improvement over the baseline is undermined.
  4. [O-MedAL Implementation Details and Table 5A] The comparison between O-MedAL and MedAL is confounded by a change of base architecture: O-MedAL uses ResNet18 while MedAL uses InceptionV3, with the paper justifying the change on the grounds that ResNet18 is smaller. Claims that O-MedAL is 'more accurate than MedAL' and that it reduces wall time by 4.5-6 times over MedAL therefore conflate the effect of the online active learning method with the effect of the architecture change. The comparison would need to hold the base architecture fixed or explicitly model architecture as a separate factor.
minor comments (5)
  1. [Abstract and Figure 1] The abstract and Eq. (2) describe the sampler as maximizing the average distance to all training examples, while Figure 1's caption and the implementation section describe selecting examples farthest from the centroid. These are different rules; the paper should state explicitly which rule was actually used in the experiments.
  2. [Table 5A, wall time] The wall-time comparisons are not directly interpretable because O-MedAL models are trained with up to 150 epochs per iteration and validation every epoch, while the ResNet18 baseline is trained for a fixed 80 epochs; the authors partially acknowledge this, but the 'Wall Time' column should not be read as a clean efficiency comparison.
  3. [Eq. (10)] Equation (10) is introduced as the cumulative number of example images used for backpropagation, but the following sentence says the cumulative number of weight updates is N_t/b; using the same symbol N_t for both images and updates is confusing and should be clarified.
  4. [Dataset Description, Messidor] The phrase 'we believe between 2005 and 2006' for Messidor collection dates is informal; it should be replaced with a citation or removed.
  5. [Figure 4 discussion] The statement that 'assuming overfitting does not decrease test accuracy' is an unverified assumption used to interpret the MedAL curve; it should be explicitly flagged as an assumption rather than implied.

Circularity Check

2 steps flagged · score 5.0 of 10

Theorem 1's proof assumes the one-dimensional rank preservation it must prove; headline accuracy is selected on the same test set.

  1. other [Theorem 1 proof, Section 'Sampling based on distance between feature embeddings', Eqs. (5)-(7)]
    "FURTHERMORE, IF z ≥ 0, THEN φRANK⟨z⟩ = φRANK⟨√z⟩ =⇒ φRANK⟨Σ√z⟩ = φRANK⟨Σz⟩. WE CAN USE THESE PROPERTIES TO SIMPLIFY BOTH SIDES OF EQ.(5) BY CONSIDERING z = f(X)^(j) AND THEN REMOVING THE SQUARE ROOT AND CONSTANT TERMS."

    The theorem must prove that ranking by Eq. (2), the average Euclidean distance to all labeled examples, is preserved by ranking by distance to their centroid. The only bridge supplied is the asserted rank-preservation of removing the square root before summing, i.e. rank⟨Σ√z⟩ = rank⟨Σz⟩. That assertion is not a consequence of scalar monotonicity; it is the one-dimensional form of the very equivalence being proved. In d=1, the theorem would claim rank(Σ_i |x_i − x|) = rank(|x − x̄|) = rank(Σ_i (x_i − x)^2), which is exactly the asserted Σ√z versus Σz step. The proof therefore assumes the conclusion instead of deriving it. The step is also false: with labeled values {0,10,10}, candidates x=6 and x=8 have average distances 14/3 and 4, but centroid distances 2/3 and 4/3, reversing the order.

  2. fitted input called prediction ['Online Active Learning (O-MedAL) Evaluation' section, Figure 5 and Table (A)]
    "WE FURTHER EVALUATE THE p = 0.125 AND p = 0.875 MODELS USING PATIENCE OF 5, 10 AND 20, AS THESE TWO VALUES OF p WERE BEST PERFORMING."

    The p/patience configuration is selected after inspecting test-accuracy curves (Figure 5C is titled 'TEST ACCURACY'), and the headline claim of 6.30% improvement and 25.29% labeling comes from the best of those selected configurations. The reported number is therefore not a held-out prediction of the method's performance; it is the maximum of a hyperparameter search evaluated on the same Messidor test set used to state the headline. Consequently the improvement is selected on the target metric, rather than independently predicted.

full rationale

The paper's only real derivation chain is Theorem 1, and that chain is broken in a question-begging way: the proof replaces sums of square roots by sums of squares using a rank property that is exactly the one-dimensional form of the target equivalence. This makes the advertised O(M) exact speedup unsupported as a derivation. Separately, some design choices (Euclidean distance, Mixed-5 layer) are justified only by the authors' prior MedAL paper, but those are configuration choices and the online method is still exercised on external datasets, so I do not treat that as the main circularity. The reported accuracy improvement is weakened by selection of p and patience on the same Messidor test set used for the headline. The experimental runs themselves are real and the method is benchmarked against random/entropy baselines, so the paper is not wholly circular; nonetheless the central speedup claim reduces to an assumed rank-preservation rather than a proof, and the headline gain is a test-set-optimized maximum.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The central design depends on the metric-space assumption for CNN features, the entropy-as-informativeness assumption, and the empirical claim that a random subset of old labels plus new labels suffices for online training. The most fragile premise is the centroid ranking identity, which is stated for Euclidean distance but only holds for squared Euclidean distance. The experimental claims add the assumption that hyperparameters transfer from Messidor to the other datasets and that the chosen baseline is well-converged.

free parameters (5)
  • p (fraction of previously labeled data reused per iteration) = 0.875 (best on Messidor)
    Introduced in Equation (10) and swept over nine values; higher p gives higher accuracy but more computation. The reported best configurations use p=0.875.
  • M (number of highest-entropy candidate samples) = 50 (Messidor and Skin), 30 (Breast)
    Set from the prior MedAL paper, not re-validated here.
  • l (images labeled per active learning iteration) = 20, 5, 10 per dataset
    Taken from the prior MedAL implementation.
  • patience and max epochs = 10 fixed epochs for most runs; patience 5, 10, or 20 with up to 150 epochs
    Early stopping parameters are selected on Messidor and drive the reported accuracy and efficiency numbers.
  • feature extraction layer = layer2 of ResNet18; Mixed6U of InceptionV3
    Chosen based on the authors' prior empirical evidence in their MedAL paper.
assumptions (6)
  • domain assumption CNN feature embeddings can be interpreted as a metric space where smaller distance implies higher visual similarity.
    Stated in 'Deep Representations as Feature Embeddings'; this is the basis for using distance in feature space as an informativeness score.
  • domain assumption Predictive entropy identifies examples close to the decision boundary and therefore informative.
    Standard active learning assumption, restated in the sampling section; used to filter candidates before distance scoring.
  • ad hoc to paper Training on a random subset of previously labeled examples plus the new labels is sufficient to retain performance.
    The paper states this is found empirically ('we have found empirically that including a non-null subset of previously labeled data is necessary') but gives no analysis of forgetting or subset composition.
  • standard math The ordering of scores is preserved when average Euclidean distance is replaced by distance to the centroid.
    This identity holds for squared Euclidean distance, not Euclidean distance; the paper's proof mistakenly removes square roots before summing. This is the load-bearing premise of Theorem 1.
  • domain assumption The validation set on Messidor transfers to the other two datasets.
    The authors select hyperparameters on Messidor and apply them to BACH and ISIC to avoid labeled validation sets for those datasets; cross-dataset transfer of hyperparameters is assumed.
  • domain assumption The random re-computation of the Messidor split preserves comparability between MedAL and O-MedAL experiments.
    The O-MedAL section describes a split of 949/238, while Table 1 uses 768/240/192; the text does not reconcile these numbers, and re-computing splits each run adds variance.

how reviews work

0 comments
Cite this review

Pith. "Pith review of O-MedAL: Online Active Deep Learning for Medical Image Analysis." pith.science (2026). https://pith.science/paper/M3BLYA7Q

@misc{pith2026190810508,
  author       = {Pith},
  title        = {Pith review of: O-MedAL: Online Active Deep Learning for Medical Image Analysis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/M3BLYA7Q}},
  note         = {Machine review of arXiv:1908.10508}
}
read the original abstract

Active Learning methods create an optimized labeled training set from unlabeled data. We introduce a novel Online Active Deep Learning method for Medical Image Analysis. We extend our MedAL active learning framework to present new results in this paper. Our novel sampling method queries the unlabeled examples that maximize the average distance to all training set examples. Our online method enhances performance of its underlying baseline deep network. These novelties contribute significant performance improvements, including improving the model's underlying deep network accuracy by 6.30%, using only 25% of the labeled dataset to achieve baseline accuracy, reducing backpropagated images during training by as much as 67%, and demonstrating robustness to class imbalance in binary and multi-class tasks.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 32 canonical work pages

  1. [1]

    write newline

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

  2. [2]

    (2004) Queries revisited

    Angluin, D. (2004) Queries revisited. Theoretical Computer Science, 313, 175--194

    work page 2004

  3. [3]

    S., Safwan, M., Alex, V., Marami, B., Prastawa, M., Chan, M., Donovan, M., Fernandez, G., Zeineh, J., Kohl, M., Walz, C., Ludwig, F., Braunewell, S., Baust, M., Vu, Q

    Aresta, G., Araújo, T., Kwok, S., Chennamsetty, S. S., Safwan, M., Alex, V., Marami, B., Prastawa, M., Chan, M., Donovan, M., Fernandez, G., Zeineh, J., Kohl, M., Walz, C., Ludwig, F., Braunewell, S., Baust, M., Vu, Q. D., To, M. N. N., Kim, E., Kwak, J. T., Galal, S., Sanchez-Freire, V., Brancati, N., Frucci, M., Riccio, D., Wang, Y., Sun, L., Ma, K., Fa...

    work page 2019

  4. [4]

    and Luz, K

    Baram, Y., El-Yaniv, R. and Luz, K. (2003) Online choice of active learning algorithms. Journal of Machine Learning Research, 5, 255--291

    work page 2003

  5. [5]

    and Zhou, J

    Cai, W., Zhang, Y. and Zhou, J. (2013) Maximizing expected model change for active learning in regression. In 2013 IEEE 13th International Conference on Data Mining, 51--60. IEEE

    work page 2013

  6. [6]

    A., Ghahramani, Z

    Cohn, D. A., Ghahramani, Z. and Jordan, M. I. (1995) Active learning with statistical models. In Advances in neural information processing systems, vol. 4, 705--712

    work page 1995

  7. [7]

    M., Smailagic , A

    Costa , P., Araújo , T., Aresta , G., Galdran , A., Mendonça , A. M., Smailagic , A. and Campilho , A. (2019) Eyewes: Weakly supervised pre-trained convolutional neural networks for diabetic retinopathy detection. In 2019 16th International Conference on Machine Vision Applications (MVA), 1--6

    work page 2019

  8. [8]

    I., Niemeijer, M., Abr \`a moff, M., Mendon c a, A

    Costa, P., Galdran, A., Meyer, M. I., Niemeijer, M., Abr \`a moff, M., Mendon c a, A. M. and Campilho, A. (2018) End-to-end adversarial retinal image synthesis. IEEE Transactions on Medical Imaging, 37, 781--791

    work page 2018

Show all 41 references
  1. [9]

    Dasgupta, S., Hsu, D. J. and Monteleoni, C. (2008) A general agnostic active learning algorithm. In Advances in neural information processing systems, 353--360

    2008

  2. [10]

    and Giles, L

    Ertekin, S., Huang, J., Bottou, L. and Giles, L. (2007) Learning on the border: Active learning in imbalanced data classification. In Proceedings of the Sixteenth ACM Conference on Conference on Information and Knowledge Management, CIKM '07, 127--136. New York, NY, USA: ACM. ...

    2007

  3. [11]

    S., Shamir, E

    Freund, Y., Seung, H. S., Shamir, E. and Tishby, N. (1993) Information, prediction, and query by committee. In Advances in neural information processing systems, 483--490

    1993

  4. [12]

    Machine learning, 28, 133--168

    --- (1997) Selective sampling using the query by committee algorithm. Machine learning, 28, 133--168

    1997

  5. [13]

    and Denzler, J

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

    2014

  6. [14]

    and Ghahramani, Z

    Gal, Y., Islam, R. and Ghahramani, Z. (2017) Deep B ayesian active learning with image data. In Proceedings of the 34th International Conference on Machine Learning (eds. D. Precup and Y. W. Teh), vol. 70 of Proceedings of Machine Learning Research, 1183--1192. International C...

    2017

  7. [15]

    and Bengio, Y

    Glorot, X. and Bengio, Y. (2010) Understanding the difficulty of training deep feedforward neural networks. In Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, 249--256

    2010

  8. [16]

    and Faure, E

    G \'o rriz, M., Gir \'o -i Nieto, X., Carlier, A. and Faure, E. (2017) Cost-effective active learning for melanoma segmentation. In ML4H: Machine Learning for Health Workshop at NIPS 2017. Long Beach, CA, USA

    2017

  9. [18]

    and Sun, J

    He, K., Zhang, X., Ren, S. and Sun, J. (2015) Deep residual learning for image recognition. CoRR, abs/1512.03385. ://arxiv.org/abs/1512.03385

    2015 arXiv

  10. [19]

    and Hahn, H

    Homeyer, A., Schenk, A., Dahmen, U., Dirsch, O., Huang, H. and Hahn, H. K. (2012) A comparison of sampling strategies for histological image analysis. Journal of Pathology Informatics, 2, 11

    2012

  11. [20]

    Kingma , D. P. and Ba , J. (2014) Adam: A Method for Stochastic Optimization . arXiv e-prints, arXiv:1412.6980

    2014 arXiv

  12. [21]

    Lewis, D. D. and Catlett, J. (1994) Heterogeneous uncertainty sampling for supervised learning. In Machine learning proceedings 1994, 148--156. Elsevier

    1994

  13. [22]

    Lewis, D. D. and Gale, W. A. (1994) A sequential algorithm for training text classifiers. In Proceedings of the 17th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 3--12. Springer-Verlag New York, Inc

    1994

  14. [23]

    E., Setio, A

    Litjens, G., Kooi, T., Bejnordi, B. E., Setio, A. A. A., Ciompi, F., Ghafoorian, M., Van Der Laak, J. A., Van Ginneken, B. and S \'a nchez, C. I. (2017) A survey on deep learning in medical image analysis. Medical image analysis, 42, 60--88

    2017

  15. [24]

    v., Maduskar, P., Philipsen, R

    Melendez, J., Ginneken, B. v., Maduskar, P., Philipsen, R. H. H. M., Ayles, H. and Sánchez, C. I. (2016) On Combining Multiple - Instance Learning and Active Learning for Computer - Aided Detection of Tuberculosis . IEEE Transactions on Medical Imaging, 35, 1013--1024

    2016

  16. [25]

    and Mooney, R

    Melville, P. and Mooney, R. J. (2004) Diverse ensembles for active learning. In Proceedings of the twenty-first international conference on Machine learning, 74. ACM

    2004

  17. [26]

    Mikolov, T., Chen, K., Corrado, G. S. and Dean, J. (2013) Efficient estimation of word representations in vector space. arXiv preprint arXiv:1605.01397. ://arxiv.org/abs/1301.3781

    2013 arXiv

  18. [27]

    and Carbonell, J

    Murugesan, K. and Carbonell, J. (2017) Active learning from peers. In Advances in Neural Information Processing Systems, 7008--7017

    2017

  19. [28]

    F., Dalca, A

    Pace, D. F., Dalca, A. V., Geva, T., Powell, A. J., Moghari, M. H. and Golland, P. (2015) Interactive Whole - Heart Segmentation in Congenital Heart Disease . In Medical Image Computing and Computer - Assisted Intervention – MICCAI 2015 , Lecture Notes in Computer Science , 80--88

    2015

  20. [29]

    W., Starr, D

    Richards, J. W., Starr, D. L., Brink, H., Miller, A. A., Bloom, J. S., Butler, N. R., James, J. B., Long, J. P. and Rice, J. (2011) ACTIVE LEARNING TO OVERCOME SAMPLE SELECTION BIAS : APPLICATION TO PHOTOMETRIC VARIABLE STAR CLASSIFICATION . The Astrophysical Journal, 744, 192...

    2011

  21. [30]

    (2007) Online active learning methods for fast label-efficient spam filtering

    Sculley, D. (2007) Online active learning methods for fast label-efficient spam filtering. In CEAS

    2007

  22. [31]

    (2008) Curious machines: Active learning with structured instances

    Settles, B. (2008) Curious machines: Active learning with structured instances. Ph.D. thesis, University of Wisconsin--Madison

    2008

  23. [32]

    Y., Walawalkar, D., Khandelwal, K., Galdran, A., Mirshekari, M., Fagert, J., Xu, S., Zhang, P

    Smailagic, A., Costa, P., Noh, H. Y., Walawalkar, D., Khandelwal, K., Galdran, A., Mirshekari, M., Fagert, J., Xu, S., Zhang, P. et al. (2018) Medal: Accurate and robust deep active learning for medical image analysis. In 2018 17th IEEE International Conference on Machine Lear...

    2018

  24. [33]

    and Wojna, Z

    Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J. and Wojna, Z. (2016) Rethinking the inception architecture for computer vision. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2818--2826

    2016

  25. [34]

    M., Hou, L., Saltz, J

    Wen, S., Kurc, T. M., Hou, L., Saltz, J. H., Gupta, R. R., Batiste, R., Zhao, T., Nguyen, V., Samaras, D. and Zhu, W. (2018) Comparison of Different Classifiers with Active Learning to Support Quality Control in Nucleus Segmentation in Pathology Images . AMIA Summits on Transl...

    2018

  26. [35]

    and Hauptmann, A

    Yang, Y., Ma, Z., Nie, F., Chang, X. and Hauptmann, A. G. (2015) Multi-class active learning by uncertainty sampling with diversity maximization. International Journal of Computer Vision, 113, 113--127

    2015

  27. [36]

    and Vinyals, O

    Zhang, C., Bengio, S., Hardt, M., Recht, B. and Vinyals, O. (2016) Understanding deep learning requires rethinking generalization. arXiv preprint arXiv:1611.03530

    2016 arXiv

  28. [37]

    Zhang , C., Tavanapong , W., Kijkul , G., Wong , J., de Groen , P. C. and Oh , J. (2018) Similarity-based active learning for image classification under class imbalance. In 2018 IEEE International Conference on Data Mining (ICDM), 1422--1427

    2018

  29. [38]

    and Gu, L

    Zhao, Y., Zhang, J., Xie, H., Zhang, S. and Gu, L. (2018) Minimization of annotation work: diagnosis of mammographic masses via active learning. Physics in Medicine and Biology, 63, 115003

    2018

  30. [39]

    and Liang, J

    Zhou, Z., Shin, J., Zhang, L., Gurudu, S., Gotway, M. and Liang, J. (2017) Fine-tuning convolutional neural networks for biomedical image analysis: actively and incrementally. In Proceedings of the IEEE conference on computer vision and pattern recognition, 7340--7351

    2017

  31. [40]

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

    ENTRY address author booktitle chapter edition editor howpublished institution journal key month note number organization pages publisher school series title type url volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.senten...

  32. [41]

    write newline

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

  33. [42]

    write newline

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

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.