Pith. sign in

REVIEW 4 major objections 4 minor 59 references

Transfer Learning-Based Label Proportions Method with Data of Uncertainty

T0 review · 4 major / 4 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 The paper proposes TL-LLP, a transfer-learning method for learning with label proportions that models each input's uncertainty as a bounded perturbation and, on nine text datasets, reports higher accuracy and lower noise sensitivity than…

desk verdict Readable transfer-LLP package whose stated noise robustness is undermined by a min-min formulation that fits training noise rather than defending against it. read the letter →

arxiv 1908.06603 v1 pith:MGKWTNXN submitted 2019-08-19 cs.LG stat.ML

Yanshan Xiao , HuaiPei Wang , Bo Liu This is my paper · ORCID
classification cs.LGstat.ML
keywords learningwithlabelproportionstransferuncertaindatasupportvectorregressionnoiserobustnessalternatingoptimizationbag-levelsupervision
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

Learning with label proportions (LLP) supplies only bag-level statistics—the fraction of positive instances in each bag—not instance labels. This paper claims that when both the source and target tasks contain uncertain or noisy data, a classifier can be improved by transferring knowledge from the source task while explicitly modeling each input as lying within a bounded 'reachability area' around the observed point. The proposed TL-LLP method decomposes the source and target weight vectors as $w_1 = w_0 + v_1$ and $w_2 = w_0 + v_2$, so a shared parameter $w_0$ carries the transferable structure, and solves an alternating SVR-style optimization that also chooses perturbation vectors $\Delta x_i$ inside their bounds. On nine text dataset pairs, the paper reports that TL-LLP achieves higher mean accuracy and smaller standard deviations than MeanMap, InvCal, alter-∝SVM, and conv-∝SVM, and that its accuracy degrades more slowly as up to 32% of training data is corrupted by Gaussian noise.

What carries the argument

The load-bearing objects are the weight decomposition $w_t = w_0 + v_t$ and the reachability area of each instance. The decomposition is what makes transfer possible: $w_0$ is the common direction learned from both tasks, and $v_1, v_2$ are small task-specific corrections, so the target classifier $f_2$ inherits structure from the source. The reachability area is the set of possible true inputs $\{x_i + \Delta x_i : \|\Delta x_i\| \le \delta_i\}$; by making $\Delta x_i$ a variable in problem (5), the model lets the optimizer choose perturbations that reduce the training slack, which is the mechanism behind the claimed noise insensitivity. Lemmas 2 and 3 are the workhorses: Lemma 2 dualizes the QP after fixing $\Delta x$, giving closed-form expressions for $w_0, v_1, v_2$; Lemma 3 solves the $\Delta x$-update in closed form using the gradient direction of the kernel, effectively pushing each noisy point back inside the $\varepsilon$-tube at cost $O(|D_1|+|D_2|)$ per iteration.

What would settle it

The paper's noise experiments corrupt the training data (Figure 5). The decisive check is to corrupt only the target test set after training, with training data clean, and compare TL-LLP with a version whose $\Delta x$ machinery is disabled; if the accuracy gap vanishes or reverses, the reported noise insensitivity is an artifact of training-time perturbation rather than genuine robustness to input uncertainty.

Watch

Extended reading notes

Core claim

The paper's central claim is that label-proportion learning can be made transferable and noise-tolerant at the same time by solving a single objective over two coupled SVR problems. For source task $T_1$ and target task $T_2$, the classifier planes are $f_1(x) = (w_0 + v_1)^T x + b_1$ and $f_2(x) = (w_0 + v_2)^T x + b_2$; the shared vector $w_0$ is what carries knowledge from source to target, while $v_1$ and $v_2$ absorb task-specific deviations. To handle uncertainty, each instance $x_i$ is assumed to be observed with additive noise $\Delta x_i$ satisfying $\|\Delta x_i\| \le \delta_i$, so the true input lies in a reachability area; the optimization (problem (5)) uses the perturbed inputs in the bag-average constraints and includes the $\Delta x_i$ as decision variables. The alternating Algorithm 1 fixes the noise vectors and solves a convex QP dual (Lemma 2) to update all hyperplane parameters, then fixes the hyperplanes and updates each $\Delta x_i$ by the closed form of Lemma 3, which moves a perturbed point toward the correct side of the classifier whenever it falls outside the $\varepsilon$-insensitive zone. The paper takes the resulting $f_2$ as the target classifier and reports, in Table 2 and Figure 5, that it beats the four baselines in accuracy and in robustness to added Gaussian noise over nine transfer dataset pairs.

Load-bearing premise

The load-bearing premise is that letting the learner move each noisy input to a nearby point that reduces its training error is a faithful model of uncertainty; if real noise does not behave that way, the claimed robustness need not transfer to test time.

Editorial extensions

If this is right

  • If TL-LLP works as reported, LLP practitioners with noisy text data can improve target-task accuracy by adding a related source task and estimating a per-sample noise bound $\delta_i$, without needing instance labels in either task.
  • The shared $w_0$ means the target classifier is not trained from scratch: the source task contributes the common decision direction, which is why the method can outperform single-task LLP baselines that see only the target bags.
  • The closed-form $\Delta x$ update in Lemma 3 keeps the added computational cost linear per iteration, so total training time stays at $m \cdot O((|D_1|+|D_2|)^2)$, dominated by the QP solve.
  • The reported lower standard deviations across bag sizes suggest that transfer plus uncertainty modeling stabilizes predictions, not just raises mean accuracy.

Reading between the lines

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

  • Extension: Lemma 3 chooses $\Delta x_i$ that shrink training slack, an optimistic perturbation; if noise is adversarial or present only at test time, the robustness claim may not carry over, since at test time the model evaluates $f_2(x)$ on the observed, unperturbed point.
  • Extension: The weight decomposition assumes the source and target tasks share a common direction $w_0$; on unrelated task pairs the transferred component could hurt, a regime the paper does not test.
  • Extension: A decisive comparison would be TL-LLP with the $\Delta x$ terms removed; this would isolate whether the reported gains come from transfer, from uncertainty modeling, or from their interaction.
  • Extension: Because the derivation uses kernel evaluations and a Taylor expansion in Lemma 3, the same transfer-and-perturb strategy extends to nonlinear kernels, although the experiments use a linear kernel.
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 / 4 minor

Summary. This paper proposes TL-LLP, a transfer learning method for learning with label proportions where both source and target data are subject to bounded additive uncertainty. The model introduces a shared parameter w0 and task-specific increments v1 and v2, and it represents uncertainty by allowing each input x to be shifted by a vector Δx with ||Δx|| ≤ δ_i. An alternating algorithm is proposed: for fixed Δx, a dual QP is solved; for fixed hyperplanes, Δx is updated via a closed-form expression. Experiments on nine text classification tasks compare TL-LLP with four LLP baselines and include a noise-sensitivity study.

Significance. If the central claims were valid, the paper would make a useful contribution by combining transfer learning with LLP under input uncertainty. The dual derivation for fixed Δx is standard, and the experimental study is fairly extensive. However, the central robustness mechanism is a minimization over the noise variables, which reduces training slack rather than providing worst-case robustness; this is a fundamental mismatch with the paper's stated claim. In addition, the experiments do not isolate the uncertainty-handling benefit from the use of additional source data. The paper includes a complete derivation of the dual and a reproducible experimental protocol, but the load-bearing noise-robustness claim is not supported by the formulation as written.

major comments (4)
  1. [Section 3.2, problem (5) and Lemma 3] The uncertainty model is not robust. For a fixed hyperplane, the slack constraints are relaxed by choosing Δx that moves each training point toward the margin, as made explicit by Lemma 3, Eq. (11). The joint optimization over (w,b,ξ,Δx) is a min-min problem that selects the most favorable perturbation for each training example. A robust treatment of bounded input uncertainty would require the constraints to hold for every Δx with ||Δx|| ≤ δ_i, i.e., a min-max formulation. Since Algorithm 1 returns only f1 and f2 and the deployed classifier has no access to Δx at test time, the training procedure fits the training noise rather than learning invariance to it. Consequently, the 'less sensitive to noise' claim in the abstract and Section 4.4 is not supported by the formulation as written.
  2. [Section 4.3 and 4.4, Tables 2 and Figure 5] The reported accuracy gains over MeanMap, InvCal, alter-∝SVM and conv-∝SVM are confounded by the fact that TL-LLP has access to source-task data while the baselines are trained only on the target task. The higher accuracy in Table 2 and the more moderate degradation in Figure 5 could be entirely due to the additional source data. An ablation that disables the Δx update (i.e., fixes Δx = 0 in the same transfer model) is required to attribute the robustness to the uncertainty mechanism. Without such a comparison, Section 4.4 does not demonstrate that TL-LLP 'reduces the effect of noises in the data.'
  3. [Equation (11) and Section 6.3] The expression 'δti−ut||ut||' in Eq. (11) and in the proof is dimensionally inconsistent (δ is a scalar, ut/||ut|| is a unit vector) and should presumably read −δti·ut/||ut||. As written, the update rule is not well-defined, which affects the core alternating step in Algorithm 1 and the reproducibility of the method.
  4. [Algorithm 1 and Section 6.3] No convergence guarantee is given for the alternating procedure. The update of Δx uses a first-order Taylor expansion of the kernel and fixes the Lagrange multipliers, so it is not shown that the objective in problem (5) decreases monotonically or that the stopping criterion reflects convergence to a stationary point. This leaves the algorithm's behavior unspecified and is a load-bearing gap for a method whose main contribution is the iterative framework.
minor comments (4)
  1. [Section 3.2, after Algorithm 1] The sentence 'For the instance x1i in target task, if f1(x1i) = w^T_2·x1i + b2 > 0' appears to use the wrong function name; it should presumably be f2(x1i) = (w0+v2)^T x1i + b2.
  2. [Lemma 3, definitions of u1 and u2] The index notation in the definitions of u1 and u2 is inconsistent, e.g., 'xmn' and 'BTm' appear in the main text while the appendix uses different target-task indices. This makes the formulas hard to follow and should be unified.
  3. [Section 4.4] The abstract claims the method is 'less sensitive to noise,' but the experiments only report accuracy at different noise percentages; a direct measure of sensitivity (e.g., the slope of accuracy versus noise level) would be more directly aligned with the claim.
  4. [Global] There are several typos, including 'bond score δi' (should be 'bound score') and 'standard derivation' (should be 'standard deviation'). References [2] and [3] are the same work in different venues and should be cross-referenced.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the derivation is a standard regularized transfer-SVR objective, and the empirical claims are tested on held-out data; the noise-robustness concern is a modeling flaw, not a circular argument.

full rationale

The TL-LLP objective (problem 5) is a regularized multi-task SVR with bag-level proportion constraints; the dual in Lemma 2 follows from a standard Lagrangian derivation, and the Delta-x update in Lemma 3 is an explicit optimization step, not a 'prediction' from fitted values. The reported accuracies in Table 2 and the noise experiments in Figure 5 are external evaluations on held-out data. Self-citations to the authors' prior work ([19,40,46,53]) appear in related work and as citations for standard techniques such as alternating optimization; they are not invoked to force the central conclusion. The strongest concern—that problem (5) minimizes over Delta-x rather than taking a worst case, so the 'less sensitive to noise' claim is not supported—is a correctness/robustness issue, not a circularity: the paper does not define robustness as the amount by which slack can be reduced, nor does it hide a fitted parameter under the name 'prediction'. No circular step can be exhibited from the paper's equations, so the score is 0.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The free parameters are standard tuning hyperparameters plus the hand-set delta_i bound. The axioms are the standard LLP scaling assumption, the regularized multi-task assumption, the questionable uncertainty-as-minimization assumption, and the experimental task-relatedness assumption. No new particles, forces, dimensions, or entities are introduced.

free parameters (4)
  • delta_i (per-sample reachability bound) = 0.01
    The bound on ||Delta x_i|| defining the uncertainty region is chosen by hand with no per-sample uncertainty estimate and no sensitivity study. Section 4.2.
  • lambda_1 and lambda_2 (task-regularization weights) = Not reported exactly; lambda_1 > lambda_2, with C1, C2 in [2^-2, 2^7] and epsilon in [0,1]
    Trade-off weights between shared and task-specific parameters are selected by tuning over a range, and the reported results depend on this choice. Sections 3.1 and 4.2.
  • C1 and C2 (margin-loss trade-offs) = Tuned in [2^-2, 2^7]
    SVM capacity hyperparameters selected by cross-validation; they are not part of the derivation but affect all experimental conclusions. Section 4.2.
  • epsilon (epsilon-insensitive zone) = Tuned in [0,1]
    Controls the tolerance in the SVR-style constraints and is chosen by hand or tuning. Section 4.2.
assumptions (4)
  • domain assumption Platt scaling y = -log(1/p - 1) maps each bag's positive proportion to a real-valued target that a linear function should predict on average.
    Taken from Rueping's InvCal and Platt's scaling, this bridge between bag proportions and real-valued targets underlies constraints (2) and (5) but is not re-derived.
  • domain assumption The source and target classifiers share a common direction w0 with task-specific deviations v1 and v2.
    This is the regularized multi-task learning assumption from Evgeniou and Pontil; it is the entire transfer mechanism and is not validated on the data.
  • ad hoc to paper Uncertainty can be represented as x + Delta x with ||Delta x|| <= delta_i, and choosing Delta x during training is a valid way to model uncertainty.
    Unlike robust SVM formulations, problem (5) minimizes over Delta x, making the training loss easier rather than enforcing worst-case robustness. This assumption is load-bearing for the noise-sensitivity claim. Section 3.1 and Lemma 3.
  • domain assumption Source and target positive classes are related because they are drawn from subcategories of the same top-level Newsgroups or Reuters category.
    The experimental construction creates a meaningful transfer relationship, but the method itself does not verify that the tasks are actually related. Section 4.2.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Transfer Learning-Based Label Proportions Method with Data of Uncertainty." pith.science (2026). https://pith.science/paper/MGKWTNXN

@misc{pith2026190806603,
  author       = {Pith},
  title        = {Pith review of: Transfer Learning-Based Label Proportions Method with Data of Uncertainty},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MGKWTNXN}},
  note         = {Machine review of arXiv:1908.06603}
}
read the original abstract

Learning with label proportions (LLP), which is a learning task that only provides unlabeled data in bags and each bag's label proportion, has widespread successful applications in practice. However, most of the existing LLP methods don't consider the knowledge transfer for uncertain data. This paper presents a transfer learning-based approach for the problem of learning with label proportions(TL-LLP) to transfer knowledge from source task to target task where both the source and target tasks contain uncertain data. Our approach first formulates objective model for the uncertain data and deals with transfer learning at the same time, and then proposes an iterative framework to build an accurate classifier for the target task. Extensive experiments have shown that the proposed TL-LLP method can obtain the better accuracies and is less sensitive to noise compared with the existing LLP methods.

Figures

Figures reproduced from arXiv: 1908.06603 by the authors.

Figure 1
Figure 1. The binary classification problem of learning with label proportions [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Illustration of reachability area of instance [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. The mean accuracy of the data sets [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (2 more)
Figure 5
Figure 5. Figure 5: Accuracy comparison among different percents of data corrupted by noise for different data [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]
Figure 6
Figure 6. Figure 6: Average computational times on different methods [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

59 extracted references · 57 canonical work pages

  1. [1]

    Learning about individuals from group statistics

    Hendrik Kuck and Nando de Freitas. Learning about individuals from group statistics. In Uncer- tainty in Artificial Intelligence (UAI) . AUAI Press, 2005

    work page 2005

  2. [2]

    Estimating labels from label proportions

    N Quadrianto, AJ Smola, TS Caetano, and QV Le. Estimating labels from label proportions. In Proceedings of the 25th International Conference on Machine Learning , pages 776–783, 2008

    work page 2008

  3. [3]

    Estimating labels from label proportions

    Novi Quadrianto, Alex J Smola, Tiberio S Caetano, and Quoc V Le. Estimating labels from label proportions. Journal of Machine Learning Research , 10(Oct):2349–2374, 2009

    work page 2009

  4. [4]

    On learning from label proportions

    Felix X Yu, Krzysztof Choromanski, Sanjiv Kumar, Tony Jebara, and Shih-Fu Chang. On learning from label proportions. arXiv preprint arXiv:1402.5902 , 2014

    arXiv 2014

  5. [5]

    Multi-class learning from class proportions

    Zilei Wang and Jiashi Feng. Multi-class learning from class proportions. Neurocomputing, 119:273– 280, 2013

    work page 2013

  6. [6]

    Modeling attributes from category-attribute proportions

    Felix X Yu, Liangliang Cao, Michele Merler, Noel Codella, Tao Chen, John R Smith, and Shih-Fu Chang. Modeling attributes from category-attribute proportions. In Proceedings of the 22nd ACM international conference on Multimedia, pages 977–980. ACM, 2014

    work page 2014

  7. [7]

    Video event detection by inferring temporal instance labels

    Kuan-Ting Lai, Felix X Yu, Ming-Syan Chen, and Shih-Fu Chang. Video event detection by inferring temporal instance labels. In Proceedings of the ieee conference on computer vision and pattern recognition, pages 2243–2250, 2014

    work page 2014

  8. [8]

    Mining the Demographics of Political Sentiment from Twitter Using Learning from Label Proportions

    Ehsan Mohammady Ardehaly and Aron Culotta. Mining the demographics of political sentiment from twitter using learning from label proportions. arXiv preprint arXiv:1708.08000 , 2017

    work page Pith review arXiv 2017

Show all 59 references
  1. [9]

    ∝svm for learning with label proportions

    Felix X Yu, Dong Liu, Sanjiv Kumar, Tony Jebara, and Shih-Fu Chang. ∝svm for learning with label proportions. arXiv preprint arXiv:1306.0886 , 2013

    2013 arXiv

  2. [10]

    Linear twin svm for learning from label proportions

    Bo Wang, Zhensong Chen, and Zhiquan Qi. Linear twin svm for learning from label proportions. In Web Intelligence and Intelligent Agent Technology (WI-IAT), 2015 IEEE/WIC/ACM International Conference on, volume 3, pages 56–59. IEEE, 2015

    2015

  3. [11]

    Learning naive bayes models for multiple-instance learning with label proportions

    Jer´ onimo Hern´ andez and I˜ naki Inza. Learning naive bayes models for multiple-instance learning with label proportions. In Conference of the Spanish Association for Artificial Intelligence , pages 134–144. Springer, 2011

    2011

  4. [12]

    Learning a generative classifier from label proportions

    Kai Fan, Hongyi Zhang, Songbai Yan, Liwei Wang, Wensheng Zhang, and Jufu Feng. Learning a generative classifier from label proportions. Neurocomputing, 139:47–55, 2014

    2014

  5. [13]

    Transfer learning by structural analogy

    Hua-Yan Wang and Qiang Yang. Transfer learning by structural analogy. In AAAI, 2011

    2011

  6. [14]

    A survey on transfer learning

    Sinno Jialin Pan, Qiang Yang, et al. A survey on transfer learning. IEEE Transactions on knowledge and data engineering , 22(10):1345–1359, 2010

    2010

  7. [15]

    Transitive transfer learning

    Ben Tan, Yangqiu Song, Erheng Zhong, and Qiang Yang. Transitive transfer learning. InProceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , pages 1155–1164. ACM, 2015. 15

    2015

  8. [16]

    Regularized multi–task learning

    Theodoros Evgeniou and Massimiliano Pontil. Regularized multi–task learning. In Tenth Acm Sigkdd International Conference on Knowledge Discovery & Data Mining , 2004

    2004

  9. [17]

    A survey of uncertain data algorithms and applications

    Charu C Aggarwal and S Yu Philip. A survey of uncertain data algorithms and applications. IEEE Transactions on Knowledge and Data Engineering , 21(5):609–623, 2009

    2009

  10. [18]

    Uncertain data clustering in distributed peer-to-peer networks

    Jin Zhou, Long Chen, CL Philip Chen, Yingxu Wang, and Han-Xiong Li. Uncertain data clustering in distributed peer-to-peer networks. IEEE transactions on neural networks and learning systems , 29(6):2392–2406, 2018

    2018

  11. [19]

    A robust one-class transfer learning method with uncertain data

    Yanshan Xiao, Bo Liu, S Yu Philip, and Zhifeng Hao. A robust one-class transfer learning method with uncertain data. Knowledge and Information Systems , 44(2):407–438, 2015

    2015

  12. [20]

    Takagi–sugeno–kang transfer learning fuzzy logic system for the adaptive recognition of epileptic electroencephalogram signals

    Changjian Yang, Zhaohong Deng, Kup-Sze Choi, and Shitong Wang. Takagi–sugeno–kang transfer learning fuzzy logic system for the adaptive recognition of epileptic electroencephalogram signals. IEEE Transactions on Fuzzy Systems , 24(5):1079–1094, 2016

    2016

  13. [21]

    Enhanced knowledge-leverage-based tsk fuzzy system modeling for inductive transfer learning

    Zhaohong Deng, Yizhang Jiang, Hisao Ishibuchi, Kup Sze Choi, and Shitong Wang. Enhanced knowledge-leverage-based tsk fuzzy system modeling for inductive transfer learning. Acm Transac- tions on Intelligent Systems & Technology , 8(1):11, 2016

    2016

  14. [22]

    Svm classifier estimation from group probabilities

    Stefan Rueping. Svm classifier estimation from group probabilities. In Proceedings of the 27th international conference on machine learning (ICML-10) , pages 911–918, 2010

    2010

  15. [23]

    Laplacian svm for learning from label proportions

    Limeng Cui, Zhensong Chen, Fan Meng, and Yong Shi. Laplacian svm for learning from label proportions. In Data Mining Workshops (ICDMW), 2016 IEEE 16th International Conference on , pages 847–852. IEEE, 2016

    2016

  16. [24]

    Learning with label proportions based on nonparallel support vector machines

    Zhensong Chen, Zhiquan Qi, Bo Wang, Limeng Cui, Fan Meng, and Yong Shi. Learning with label proportions based on nonparallel support vector machines. Knowledge-Based Systems, 119:126–141, 2017

    2017

  17. [25]

    Learning with label proportions via npsvm

    Zhiquan Qi, Bo Wang, Fan Meng, and Lingfeng Niu. Learning with label proportions via npsvm. IEEE transactions on cybernetics , 47(10):3293–3305, 2017

    2017

  18. [26]

    A probabilistic approach for learning with label pro- portions applied to the us presidential election

    Tao Sun, Dan Sheldon, and Brendan OConnor. A probabilistic approach for learning with label pro- portions applied to the us presidential election. In Data Mining (ICDM), 2017 IEEE International Conference on, pages 445–454. IEEE, 2017

    2017

  19. [27]

    Learning from label proportions by optimizing cluster model se- lection

    Marco Stolpe and Katharina Morik. Learning from label proportions by optimizing cluster model se- lection. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases, pages 349–364. Springer, 2011

    2011

  20. [28]

    On the complexity of learning from label proportions

    Benjamin Fish and Lev Reyzin. On the complexity of learning from label proportions. In Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence, IJCAI , pages 19–25, 2017

    2017

  21. [29]

    Learning from label proportions on high- dimensional data

    Yong Shi, Jiabin Liu, Zhiquan Qi, and Bo Wang. Learning from label proportions on high- dimensional data. Neural Networks, 103:9–18, 2018

    2018

  22. [30]

    Hierarchical density-based clustering of uncertain data

    H-P Kriegel and Martin Pfeifle. Hierarchical density-based clustering of uncertain data. In Data Mining, Fifth IEEE International Conference on , pages 4–pp. IEEE, 2005

    2005

  23. [31]

    Effi- cient clustering of uncertain data

    Wang Kay Ngai, Ben Kao, Chun Kit Chui, Reynold Cheng, Michael Chau, and Kevin Y Yip. Effi- cient clustering of uncertain data. In Data Mining, 2006. ICDM’06. Sixth International Conference on, pages 436–445. IEEE, 2006

    2006

  24. [32]

    On density based transforms for uncertain data mining

    Charu C Aggarwal. On density based transforms for uncertain data mining. In Data Engineering,

  25. [33]

    Large margin clustering on uncertain data by considering probability distribution similarity

    Lei Xu, Qinghua Hu, Edward Hung, Baowen Chen, Xu Tan, and Changrui Liao. Large margin clustering on uncertain data by considering probability distribution similarity. Neurocomputing, 158:81–89, 2015. 16

    2015

  26. [34]

    Support vector classification with input data uncertainty

    Jinbo Bi and Tong Zhang. Support vector classification with input data uncertainty. In Advances in neural information processing systems , pages 161–168, 2005

    2005

  27. [35]

    Direct mining of discriminative patterns for classifying un- certain data

    Chuancong Gao and Jianyong Wang. Direct mining of discriminative patterns for classifying un- certain data. In Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining , pages 861–870. ACM, 2010

    2010

  28. [36]

    Decision trees for uncertain data

    Smith Tsang, Ben Kao, Kevin Y Yip, Wai-Shing Ho, and Sau Dan Lee. Decision trees for uncertain data. IEEE transactions on knowledge and data engineering , 23(1):64–78, 2011

    2011

  29. [37]

    Classification of uncertain data streams based on extreme learning machine

    Keyan Cao, Guoren Wang, Donghong Han, Jingwei Ning, and Xin Zhang. Classification of uncertain data streams based on extreme learning machine. Cognitive Computation, 7(1):150–160, 2015

    2015

  30. [38]

    Two birds with one stone: Classifying positive and unlabeled examples on uncertain data streams

    Donghong Han, Shuoru Li, Fulin Wei, Yuying Tang, Feida Zhu, and Guoren Wang. Two birds with one stone: Classifying positive and unlabeled examples on uncertain data streams. Neurocomputing, 277:149–160, 2018

    2018

  31. [39]

    Maximum likelihood estimation from uncertain data in the belief function frame- work

    Thierry Denoeux. Maximum likelihood estimation from uncertain data in the belief function frame- work. IEEE Transactions on knowledge and data engineering , 25(1):119–130, 2013

    2013

  32. [40]

    Svdd-based outlier detection on uncertain data

    Bo Liu, Yanshan Xiao, Longbing Cao, Zhifeng Hao, and Feiqi Deng. Svdd-based outlier detection on uncertain data. Knowledge and information systems , 34(3):597–618, 2013

    2013

  33. [41]

    A novel anomaly detection algorithm for sensor data under uncertainty

    Raihan Ul Islam, Mohammad Shahadat Hossain, and Karl Andersson. A novel anomaly detection algorithm for sensor data under uncertainty. Soft Computing, 22(5):1623–1639, 2018

    2018

  34. [42]

    A simple method for estimating conditional probabilities for svms

    Stefan R¨ uping. A simple method for estimating conditional probabilities for svms. Technical re- port, Technical Report/Universit¨ at Dortmund, SFB 475 Komplexit¨ atsreduktion in Multivariaten Datenstrukturen, 2004

    2004

  35. [43]

    Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods

    John Platt et al. Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods. Advances in large margin classifiers , 10(3):61–74, 1999

    1999

  36. [44]

    The total least squares problem: Computational aspects and analysis

    Sabine Van Huffel and Joos Vandewalle. The total least squares problem: Computational aspects and analysis. Mathematics of Computation , 59(200):724, 1991

    1991

  37. [45]

    Support vector machines with input data uncertainty

    J Bi and T Zhang. Support vector machines with input data uncertainty. Proc. Advances in Neural Information Processing Systems, 2004 , 2004

    2004

  38. [46]

    Uncertain one- class learning and concept summarization learning on uncertain data streams

    Bo Liu, Yanshan Xiao, S Yu Philip, Longbing Cao, Yun Zhang, and Zhifeng Hao. Uncertain one- class learning and concept summarization learning on uncertain data streams. IEEE Transactions on Knowledge and Data Engineering , 26(2):468–484, 2014

    2014

  39. [47]

    Statistical learning theory

    Edwin PD Pednault. Statistical learning theory. Citeseer, 1997

    1997

  40. [48]

    Linear time maximum margin clustering

    Fei Wang, Bin Zhao, and Changshui Zhang. Linear time maximum margin clustering. IEEE Transactions on Neural Networks , 21(2):319–332, 2010

    2010

  41. [49]

    Transferring naive bayes classifiers for text classification

    Wenyuan Dai, Gui-Rong Xue, Qiang Yang, and Yong Yu. Transferring naive bayes classifiers for text classification. In AAAI, volume 7, pages 540–545, 2007

    2007

  42. [50]

    Cross domain distribution adaptation via kernel mapping

    Erheng Zhong, Wei Fan, Jing Peng, Kun Zhang, Jiangtao Ren, Deepak Turaga, and Olivier Ver- scheure. Cross domain distribution adaptation via kernel mapping. In Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 1027–1036. ...

    2009

  43. [51]

    M3miml: A maximum margin method for multi-instance multi- label learning

    Min-Ling Zhang and Zhi-Hua Zhou. M3miml: A maximum margin method for multi-instance multi- label learning. In 2008 Eighth IEEE International Conference on Data Mining, pages 688–697. IEEE, 2008

    2008

  44. [52]

    Multi-instance learning by treating instances as non- iid samples

    Zhi-Hua Zhou, Yu-Yin Sun, and Yu-Feng Li. Multi-instance learning by treating instances as non- iid samples. In Proceedings of the 26th annual international conference on machine learning , pages 1249–1256. ACM, 2009. 17

    2009

  45. [53]

    A selective multiple instance transfer learning method for text categorization problems

    Bo Liu, Yanshan Xiao, and Zhifeng Hao. A selective multiple instance transfer learning method for text categorization problems. Knowledge-Based Systems, 141:178–187, 2018

    2018

  46. [54]

    Machine learning in automated text categorization

    Fabrizio Sebastiani. Machine learning in automated text categorization. ACM computing surveys (CSUR), 34(1):1–47, 2002

    2002

  47. [55]

    Outlier detection with uncertain data

    Charu C Aggarwal and Philip S Yu. Outlier detection with uncertain data. In Proceedings of the 2008 SIAM International Conference on Data Mining , pages 483–493. SIAM, 2008

    2008

  48. [56]

    An efficient approach for outlier detection with imperfect data labels

    Bo Liu, Yanshan Xiao, S Yu Philip, Zhifeng Hao, and Longbing Cao. An efficient approach for outlier detection with imperfect data labels. IEEE transactions on knowledge and data engineering , 26(7):1602–1616, 2014

    2014

  49. [57]

    Calculus: An intuitive and physical approach

    Morris Kline. Calculus: An intuitive and physical approach . Courier Corporation, 1998

    1998

  50. [58]

    A survey on cauchy-bunyakovsky-schwarz type discrete inequalities

    Sever S Dragomir. A survey on cauchy-bunyakovsky-schwarz type discrete inequalities. J. Inequal. Pure Appl. Math , 4(3):1–142, 2003. 18

    2003

  51. [2007]

    IEEE 23rd International Conference on , pages 866–875

    ICDE 2007. IEEE 23rd International Conference on , pages 866–875. IEEE, 2007

    2007

Pith tools

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