Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

awesome-imbalanced-learning

Everything about class-imbalanced/long-tail learning: papers, codes, frameworks, and libraries | 有关类别不平衡/长尾学习的一切：论文、代码、框架与库
https://github.com/ZhiningLiu1998/awesome-imbalanced-learning

Last synced: 3 days ago
JSON representation

  • 2.1 Surveys

  • 2.2 Ensemble Learning

    • [**Paper**
    • [**Code**
    • [**Code**
    • [**Paper** - learn/scikit-learn/blob/95d4f0841/sklearn/ensemble/_weight_boosting.py#L285)]** - Adaptive Boosting with C4.5
    • [**Paper** - Boosting with Data Generation for Imbalanced Data
    • [**Paper** - ensemble/blob/main/imbalanced_ensemble/ensemble/over_sampling/smote_bagging.py)]** - Synthetic Minority Over-sampling TEchnique Boosting
    • [**Paper** - Modified Synthetic Minority Over-sampling TEchnique Boosting
    • [**Paper** - algorithms/blob/master/ramo.py#L133)]** - Ranked Minority Over-sampling in Boosting
    • [**Paper** - ensemble/blob/main/imbalanced_ensemble/ensemble/under_sampling/rus_boost.py)]** - Random Under-Sampling Boosting
    • [**Paper** - Adaptive Boosting with Negative Correlation Learning
    • [**Paper** - Evolutionary Under-sampling in Boosting
    • [**Paper** - learn/scikit-learn/blob/95d4f0841/sklearn/ensemble/_bagging.py#L433)]** - Bagging predictor
    • [**Paper**
    • [**Code**
    • [**Code**
    • [**Code**
    • [**Paper** - ensemble/blob/main/imbalanced_ensemble/ensemble/reweighting/adacost.py)]** - Misclassification Cost-sensitive boosting
    • [**Paper** - ensemble/blob/main/imbalanced_ensemble/ensemble/reweighting/adauboost.py)]** - AdaBoost with Unequal loss functions
    • [**Paper** - ensemble/blob/main/imbalanced_ensemble/ensemble/reweighting/asymmetric_boost.py)]** - Asymmetric AdaBoost and detector cascade
    • [**Paper** - paced-ensemble)][[**Slides**](https://zhiningliu.com/files/ICDE_2020_self_paced_ensemble_slides.pdf)][[**Zhihu/知乎**](https://zhuanlan.zhihu.com/p/86891438)][[**PyPI**](https://pypi.org/project/self-paced-ensemble/)]**
    • [**Paper** - paced-ensemble)][[**Slides**](https://zhiningliu.com/files/ICDE_2020_self_paced_ensemble_slides.pdf)][[**Zhihu/知乎**](https://zhuanlan.zhihu.com/p/86891438)][[**PyPI**](https://pypi.org/project/self-paced-ensemble/)]**

  • 2.3 Data resampling

    • [**Code** - Random Over-sampling
    • [**Paper** - learn-contrib/imbalanced-learn/blob/master/imblearn/over_sampling/_smote.py#L36)]** - Synthetic Minority Over-sampling TEchnique
    • [**Paper** - learn-contrib/imbalanced-learn/blob/master/imblearn/over_sampling/_smote.py#L220)]** - Borderline-Synthetic Minority Over-sampling TEchnique
    • [**Paper** - learn-contrib/imbalanced-learn/blob/master/imblearn/over_sampling/_adasyn.py)]** - ADAptive SYNthetic Sampling
    • [**Paper** - Selective Preprocessing of Imbalanced Data
    • [**Paper** - Mahalanobis Distance-based Over-sampling for *Multi-Class* imbalanced problems.
    • [**Code** - Random Under-sampling
    • [**Paper** - learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_condensed_nearest_neighbour.py)]** - Condensed Nearest Neighbor
    • [**Paper** - learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_edited_nearest_neighbours.py)]** - Edited Condensed Nearest Neighbor
    • [**Paper** - learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_tomek_links.py)]** - Tomek's modification of Condensed Nearest Neighbor
    • [**Paper** - learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_neighbourhood_cleaning_rule.py)]** - Neighborhood Cleaning Rule
    • [**Paper** - learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_nearmiss.py)]** - Several kNN approaches to unbalanced data distributions.
    • [**Paper** - Condensed Nearest Neighbor + TomekLink
    • [**Paper** - learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_one_sided_selection.py)]** - One Side Selection
    • [**Paper** - Evolutionary Under-sampling
    • [**Paper** - learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_instance_hardness_threshold.py)]** - Instance Hardness Threshold
    • [**Paper**
    • [**Code**
    • [**Code**
    • [**Paper** - variants.readthedocs.io/en/latest/_modules/smote_variants/_smote_variants.html#SMOTE_RSB)]** - Hybrid Preprocessing using SMOTE and Rough Sets Theory
    • [**Paper** - variants.readthedocs.io/en/latest/_modules/smote_variants/_smote_variants.html#SMOTE_IPF)]** - SMOTE with Iterative-Partitioning Filter
    • **smote-variants**
    • [**Paper** - Evolutionary Under-sampling
    • [**Paper** - learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_tomek_links.py)]** - Tomek's modification of Condensed Nearest Neighbor
    • [**Paper** - learn-contrib/imbalanced-learn/blob/master/imblearn/over_sampling/_smote.py#L36)]** - Synthetic Minority Over-sampling TEchnique
    • [**Paper** - learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_instance_hardness_threshold.py)]** - Instance Hardness Threshold

  • 2.4 Cost-sensitive Learning

    • [**Paper** - Cost-sensitive SVMs for highly imbalanced classification
    • [**Paper** - Training cost-sensitive neural networks with methods addressing the class imbalance problem.
    • [**Paper** - An instance-weighting method to induce cost-sensitive trees
    • [**Paper** - An instance-weighting method to induce cost-sensitive trees
    • [**Paper** - An instance-weighting method to induce cost-sensitive trees

  • 2.5 Deep Learning

    • [**Paper**
    • [**Paper**
    • [**Paper**
    • [**Paper** - Song/TAM)]
    • [**Paper**
    • [**Paper**
    • [**Paper** - Park/GraphENS)]
    • [**Paper**
    • [**Paper**
    • [**Paper** - In the later phase of NN training, only do gradient back-propagation for "hard examples" (i.e., with large loss value)
    • [**Paper** - A uniform loss function that focuses training on a sparse set of hard examples to prevents the vast number of easy negatives from overwhelming the detector during training.
    • [**Paper** - Mean (square) false error that can equally capture classification errors from both the majority class and the minority class.
    • [**Paper**
    • [**Paper** - Class Rectification Loss for minimizing the dominant effect of majority classes by discovering sparsely sampled boundaries of minority classes in an iterative batch-wise learning process.
    • [**Paper** - DRW)] - A theoretically-principled label-distribution-aware margin (LDAM) loss motivated by minimizing a margin-based generalization bound.
    • [**Paper** - Compared to Focal Loss, which only down-weights "easy" negative examples, GHM also down-weights "very hard" examples as they are likely to be outliers.
    • [**Paper** - balanced-loss)] - a simple and generic class-reweighting mechanism based on Effective Number of Samples.
    • [**Paper** - Loss)]
    • [**Paper**
    • [**Paper** - Loss)]
    • [**Paper** - research/learning-to-reweight-examples)] - Implicitly learn a weight function to reweight the samples in gradient updates of DNN.
    • [**Paper** - weight-net)] - Explicitly learn a weight function (with an MLP as the function approximator) to reweight the samples in gradient updates of DNN.
    • [**Paper** - data-manipulation)]
    • [**Paper** - lee/l2b)]
    • [**Paper**
    • [**Paper**
    • [**Paper** - balancing)]
    • [**Paper**
    • [**Paper** - Rotation)]
    • [**Paper** - RIPL/UNO-IC)]
    • [**Paper** - research/google-research/tree/master/logit_adjustment)]
    • [**Paper** - semi-self)][[**Video**](https://www.youtube.com/watch?v=XltXZ3OZvyI&feature=youtu.be)]
    • [**Paper**
    • [**Paper**
    • [**Paper**
    • [**Paper** - oh/daso)]
    • [**Paper**
    • [**Paper**
    • [**Paper** - Nanjing/BBN)]
    • [**Paper**
    • [**Paper**
    • [**Paper** - imbalanced-regression)]
    • [**Paper** - ren/BalancedMSE)]
    • [**Paper** - regression)][[**Video**](https://www.youtube.com/watch?v=grJGixofQRU)]
    • J - based-weighting-for-imbalanced-regression)]
    • [**Paper**
    • [**Paper**
    • [**Paper**
    • [**Paper**
    • [**Paper**
    • [**Paper** - ->
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • [**Paper** - balancing)]
    • [**Paper** - semi-self)][[**Video**](https://www.youtube.com/watch?v=XltXZ3OZvyI&feature=youtu.be)]
    • [**Paper** - Nanjing/BBN)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • [**Paper** - regression)][[**Video**](https://www.youtube.com/watch?v=grJGixofQRU)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • [**Paper** - effective-ensemble-imbal?s=d3745afc-cfcf-4d60-9f34-63d3d811b55f)]
    • [**Paper** - oh/daso)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • [**Paper** - Class Rectification Loss for minimizing the dominant effect of majority classes by discovering sparsely sampled boundaries of minority classes in an iterative batch-wise learning process.
    • J - based-weighting-for-imbalanced-regression)]
    • J - based-weighting-for-imbalanced-regression)]
    • [**Paper** - research/google-research/tree/master/logit_adjustment)]
    • J - based-weighting-for-imbalanced-regression)]
    • [**Paper** - effective-ensemble-imbal?s=d3745afc-cfcf-4d60-9f34-63d3d811b55f)]
    • J - based-weighting-for-imbalanced-regression)]
    • [**Paper** - RIPL/UNO-IC)]
    • J - based-weighting-for-imbalanced-regression)]
    • [**Paper**
    • [**Paper**
    • [**Paper** - Compared to Focal Loss, which only down-weights "easy" negative examples, GHM also down-weights "very hard" examples as they are likely to be outliers.
    • [**Paper**
    • [**Paper**
    • J - based-weighting-for-imbalanced-regression)]

  • 3.1 Datasets

  • 3.2 Github Repositories

  • Uncategorized

    • Uncategorized

      • **imbalanced-ensemble** - ensemble)][[**Documentation**](https://imbalanced-ensemble.readthedocs.io/)][[**Gallery**](https://imbalanced-ensemble.readthedocs.io/en/latest/auto_examples/index.html#)][[**Paper**](https://arxiv.org/pdf/2111.12776.pdf)]
      • **imbalanced-learn** - learn-contrib/imbalanced-learn)][[**Documentation**](https://imbalanced-learn.org/stable/)][[**Paper**](https://www.jmlr.org/papers/volume18/16-365/16-365.pdf)]
      • **smote_variants** - variants.readthedocs.io/en/latest/)][[**Github**](https://github.com/analyticalmindsltd/smote_variants)] - A collection of 85 minority ***over-sampling*** techniques for imbalanced learning with multi-class oversampling and model selection features (All writen in Python, also support R and Julia).
      • **caret** - Contains the implementation of Random under/over-sampling.
      • **ROSE** - 3)] - Contains the implementation of [ROSE](https://journal.r-project.org/archive/2014-1/menardi-lunardon-torelli.pdf) (Random Over-Sampling Examples).
      • **DMwR** - Contains the implementation of [SMOTE](https://arxiv.org/pdf/1106.1813.pdf) (Synthetic Minority Over-sampling TEchnique).
      • **KEEL** - SoftComputing-Keel1.0.pdf)] - KEEL provides a simple ***GUI based*** on data flow to design experiments with different datasets and computational intelligence algorithms (***paying special attention to evolutionary algorithms***) in order to assess the behavior of the algorithms. This tool includes many widely used imbalanced learning techniques such as (evolutionary) over/under-resampling, cost-sensitive learning, algorithm modification, and ensemble learning methods.
      • API reference
      • imbalanced-ensemble - ensemble)][[Documentation](https://imbalanced-ensemble.readthedocs.io/)].
      • joblib
      • **undersampling** - A Scala library for ***under-sampling and their ensemble variants*** in imbalanced classification.
      • scikit-learn - learn-contrib](https://github.com/scikit-learn-contrib) projects.
Programming Languages
Python 4 Jupyter Notebook 3 HTML 2 Scala 1
Categories
2.5 Deep Learning 125 2.3 Data resampling 26 2.2 Ensemble Learning 21 Uncategorized 12 3.2 Github Repositories 11 2.4 Cost-sensitive Learning 5 2.1 Surveys 5 3.1 Datasets 1
Sub Categories
Uncategorized 12 3.2.1 *Algorithms & Utilities & Jupyter Notebooks* 6 3.2.3 *Slides* 3 3.2.2 *Paper list* 2
Keywords
imbalanced-data 3 machine-learning 3 classification 2 python 2 undersampling 1 nearest-neighbor-rules 1 imbalance-learning 1 algorithm 1 threading 1 parallel-computing 1 multiprocessing 1 memoization 1 caching 1 recognition 1 opensource-management 1 opensource 1 open-source-tooling 1 contributors 1 all-contributors 1 acknowledgements 1 unsupervised-learning 1 time-series-analysis 1 outlier-ensembles 1 outlier-detection 1 outlier 1 graph-neural-networks 1 fraud-detection 1 fraud 1 data-mining 1 awesome-list 1 awesome 1 anomaly-detection 1 tutorial 1 pytorch 1 image-classification 1 data-sampling 1 notre-dame 1 data-science 1