多类别不平衡分类-解决方案：概述【集成学习、数据重采样、深度学习(元学习)、异常检测】

朴灿烈づ我的快乐病毒、 2023-10-08 23:47 18阅读 0赞

**类别不平衡**（又称**长尾问题**）是指在分类问题中，类别之间的表示质量/样本数量不平等。

类别不平衡在实践中广泛存在，例如金融欺诈检测、入侵检测、医疗辅助诊断等罕见模式识别任务。

类的不平衡往往会导致传统机器学习算法的预测性能下降。类别不平衡学习旨在解决这一问题，即从不平衡的数据中学习一个无偏的预测模型。

## 目录 ##

*  [目录][Link 1]
 *  [框架与库 | Frameworks and Libraries][_ Frameworks and Libraries]
    
     *  [Python][]
     *  [R][]
     *  [Java][]
     *  [Scalar][]
     *  [Julia][]
 *  [研究论文 | Research Papers][_ Research Papers]
    
     *  [综述 | Surveys][_ Surveys]
     *  [集成学习 | Ensemble Learning][_ Ensemble Learning]
        
         *  [通用集成框架 | General ensemble][_ General ensemble]
         *  [基于 Boosting 的方法 | Boosting-based][Boosting _ _ Boosting-based]
         *  [基于 Bagging 的方法 | Bagging-based][Bagging _ _ Bagging-based]
         *  [基于代价敏感学习的方法 | Cost-sensitive ensemble][_ Cost-sensitive ensemble]
     *  [数据重采样 | Data resampling][_ Data resampling]
        
         *  [过采样 | Over-sampling][_ Over-sampling]
         *  [欠采样 | Under-sampling][_ Under-sampling]
         *  [混合采样 | Hybrid-sampling][_ Hybrid-sampling]
     *  [代价敏感学习 | Cost-sensitive Learning][_ Cost-sensitive Learning]
     *  [深度学习 | Deep Learning][_ Deep Learning]
        
         *  [综述 | Surveys][_ Surveys 1]
         *  [图数据挖掘 | Graph Neural Networks][_ Graph Neural Networks]
         *  [难例挖掘 | Hard example mining][_ Hard example mining]
         *  [损失函数设计 | Loss function engineering][_ Loss function engineering]
         *  [元学习 | Meta-learning][_ Meta-learning]
         *  [表示学习 | Representation Learning][_ Representation Learning]
         *  [后验概率校准 | Posterior Recalibration][_ Posterior Recalibration]
         *  [半监督/自监督学习 | Semi/Self-supervised Learning][_ Semi_Self-supervised Learning]
         *  [课程学习 | Curriculum Learning][_ Curriculum Learning]
         *  [双阶段训练 | Two-phase Training][_ Two-phase Training]
         *  [网络结构 | Network Architecture][_ Network Architecture]
         *  [深度生成网络 | Deep Generative Model][_ Deep Generative Model]
         *  [不平衡回归 | Imbalanced Regression][_ Imbalanced Regression]
     *  [异常检测 | Anomaly Detection][_ Anomaly Detection]
 *  [杂项 | Miscellaneous][_ Miscellaneous]
    
     *  [数据集 | Datasets][_ Datasets]
     *  [Github 项目 | Github Repositories][Github _ _ Github Repositories]
        
         *  [算法实现 & 实用程序 & 教程 | Algorithms & Utilities & Jupyter Notebooks][_ _ _ _ _ Algorithms _ Utilities _ Jupyter Notebooks]
         *  [论文列表 | Paper list][_ Paper list]
         *  [幻灯片 | Slides][_ Slides]
 *  [Contributors ✨][Contributors]

## 框架与库 | Frameworks and Libraries ##

#### Python ####

*  [imbalanced-ensemble][] \[[Github][]\]\[[Documentation][imbalanced-ensemble]\]\[[Gallery][]\]\[[Paper][]\]
    
    > **NOTE:** written in python, easy to use.
    
     *  `imbalanced-ensemble` is a Python toolbox for quick implementing and deploying ***ensemble learning algorithms*** on class-imbalanced data. It is featured for:
        
         *  (i) Unified, easy-to-use APIs, detailed [documentation][imbalanced-ensemble] and [examples][Gallery].
         *  (ii) Capable for multi-class imbalanced learning out-of-box.
         *  (iii) Optimized performance with parallelization when possible using [joblib][].
         *  (iv) Powerful, customizable, interactive training logging and visualizer.
         *  (v) Full compatibility with other popular packages like [scikit-learn][] and [imbalanced-learn][].
     *  Currently (v0.1.4), it includes more than 15 ensemble algorithms based on ***re-sampling*** and ***cost-sensitive learning*** (e.g., *SMOTEBoost/Bagging, RUSBoost/Bagging, AdaCost, EasyEnsemble, BalanceCascade, SelfPacedEnsemble*, ...).
 *  [imbalanced-learn][] \[[Github][Github 1]\]\[[Documentation][imbalanced-learn]\]\[[Paper][Paper 1]\]
    
    > **NOTE:** written in python, easy to use.
    
     *  `imbalanced-learn` is a python package offering a number of ***re-sampling*** techniques commonly used in datasets showing strong between-class imbalance. It is compatible with [scikit-learn][] and is part of [scikit-learn-contrib][] projects.
     *  Currently (v0.8.0), it includes 21 different re-sampling techniques, including over-sampling, under-sampling and hybrid ones (e.g., *SMOTE, ADASYN, TomekLinks, NearMiss, OneSideSelection*, SMOTETomek, ...)
     *  This package also provides many utilities, e.g., *Batch generator for Keras/TensorFlow*, see [API reference][].
 *  [smote\_variants][smote_variants] \[[Documentation][smote_variants]\]\[[Github][Github 2]\] - 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).

#### R ####

*  [smote\_variants][smote_variants] \[[Documentation][smote_variants]\]\[[Github][Github 2]\] - 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][] \[[Documentation][]\]\[[Github][Github 3]\] - Contains the implementation of Random under/over-sampling.
 *  [ROSE][] \[[Documentation][Documentation 1]\] - Contains the implementation of [ROSE][ROSE 1] (Random Over-Sampling Examples).
 *  [DMwR][] \[[Documentation][Documentation 2]\] - Contains the implementation of [SMOTE][] (Synthetic Minority Over-sampling TEchnique).

#### Java ####

*  [KEEL][] \[[Github][Github 4]\]\[[Paper][Paper 2]\] - 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.
    
    > **NOTE:** wide variety of classical classification, regression, preprocessing algorithms included.

#### Scalar ####

*  [undersampling][] \[[Documentation][Documentation 3]\]\[[Github][undersampling]\] - A Scala library for ***under-sampling and their ensemble variants*** in imbalanced classification.

#### Julia ####

## 研究论文 | Research Papers ##

### 综述 | Surveys ###

*  **Learning from imbalanced data (IEEE TKDE, 2009, 6000+ citations) \[[Paper][Paper 3]\]**
    
     *  Highly cited, classic survey paper. It systematically reviewed the popular solutions, evaluation metrics, and challenging problems in future research in this area (as of 2009).
 *  **Learning from imbalanced data: open challenges and future directions (2016, 900+ citations) \[[Paper][Paper 4]\]**
    
     *  This paper concentrates on the open issues and challenges in imbalanced learning, i.e., extreme class imbalance, imbalance in online/stream learning, multi-class imbalanced learning, and semi/un-supervised imbalanced learning.
 *  **Learning from class-imbalanced data: Review of methods and applications (2017, 900+ citations) \[[Paper][Paper 5]\]**
    
     *  A recent exhaustive survey of imbalanced learning methods and applications, a total of 527 papers were included in this study. It provides several detailed taxonomies of existing methods and also the recent trend of this research area.

### 集成学习 | Ensemble Learning ###

通用集成框架 | *General ensemble*

*  **Self-paced Ensemble (ICDE 2020, 20+ citations) \[[Paper][Paper 6]\]\[[Code][]\]\[[Slides][]\]\[[Zhihu/知乎][Zhihu]\]\[[PyPI][]\]**
    
    > **NOTE:** versatile solution with outstanding performance and computational efficiency.
 *  **MESA: Boost Ensemble Imbalanced Learning with MEta-SAmpler (NeurIPS 2020) \[[Paper][Paper 7]\]\[[Code][Code 1]\]\[[Video][]\]\[[Zhihu/知乎][Zhihu 1]\]**
    
    > **NOTE:** learning an optimal sampling policy directly from data.
 *  **Exploratory Undersampling for Class-Imbalance Learning (IEEE Trans. on SMC, 2008, 1300+ citations) \[[Paper][Paper 8]\]**
    
    > **NOTE:** simple but effective solution.
    
     *  EasyEnsemble \[[Code][Code 2]\]
     *  BalanceCascade \[[Code][Code 3]\]

基于 Boosting 的方法 | *Boosting-based*

*  **AdaBoost (1995, 18700+ citations) \[[Paper][Paper 9]\]\[[Code][Code 4]\]** \- Adaptive Boosting with C4.5
 *  **DataBoost (2004, 570+ citations) \[[Paper][Paper 10]\]** \- Boosting with Data Generation for Imbalanced Data
 *  **SMOTEBoost (2003, 1100+ citations) [\[Paper\]][Paper 11]\[[Code][Code 5]\]** \- Synthetic Minority Over-sampling TEchnique Boosting
 *  **MSMOTEBoost (2011, 1300+ citations) \[[Paper][Paper 12]\]** \- Modified Synthetic Minority Over-sampling TEchnique Boosting
 *  **RAMOBoost (2010, 140+ citations) \[[Paper][Paper 13]\] \[[Code][Code 6]\]** \- Ranked Minority Over-sampling in Boosting
 *  **RUSBoost (2009, 850+ citations) \[[Paper][Paper 14]\] \[[Code][Code 7]\]** \- Random Under-Sampling Boosting
 *  **AdaBoostNC (2012, 350+ citations) \[[Paper][Paper 15]\]** \- Adaptive Boosting with Negative Correlation Learning
 *  **EUSBoost (2013, 210+ citations) \[[Paper][Paper 16]\]** \- Evolutionary Under-sampling in Boosting

基于 Bagging 的方法 | *Bagging-based*

*  **Bagging (1996, 20000+ citations) \[[Paper][Paper 17]\]\[[Code][Code 8]\]** \- Bagging predictor
 *  **Diversity Analysis on Imbalanced Data Sets by Using Ensemble Models (2009, 400+ citations) \[[Paper][Paper 18]\]**
    
     *  **UnderBagging** \[[Code][Code 9]\]
     *  **OverBagging** \[[Code][Code 10]\]
     *  **SMOTEBagging** \[[Code][Code 5]\]

基于代价敏感学习的方法 | *Cost-sensitive ensemble*

*  **AdaCost (ICML 1999, 800+ citations) \[[Paper][Paper 19]\]\[[Code][Code 11]\]** \- Misclassification Cost-sensitive boosting
 *  **AdaUBoost (NIPS 1999, 100+ citations) \[[Paper][Paper 20]\]\[[Code][Code 12]\]** \- AdaBoost with Unequal loss functions
 *  **AsymBoost (NIPS 2001, 700+ citations) \[[Paper][Paper 21]\]\[[Code][Code 13]\]** \- Asymmetric AdaBoost and detector cascade

### 数据重采样 | Data resampling ###

过采样 | *Over-sampling*

*  **ROS \[[Code][Code 14]\]** \- Random Over-sampling
 *  **SMOTE (2002, 9800+ citations) \[[Paper][SMOTE]\]\[[Code][Code 15]\]** \- Synthetic Minority Over-sampling TEchnique
 *  **Borderline-SMOTE (2005, 1400+ citations) \[[Paper][Paper 22]\]\[[Code][Code 16]\]** \- Borderline-Synthetic Minority Over-sampling TEchnique
 *  **ADASYN (2008, 1100+ citations) \[[Paper][Paper 23]\]\[[Code][Code 17]\]** \- ADAptive SYNthetic Sampling
 *  **SPIDER (2008, 150+ citations) \[[Paper][Paper 24]\]\[[Code(Java)][Code_Java]\]** \- Selective Preprocessing of Imbalanced Data
 *  **Safe-Level-SMOTE (2009, 370+ citations) \[[Paper][Paper 25]\]\[[Code(Java)][Code_Java 1]\]** \- Safe Level Synthetic Minority Over-sampling TEchnique
 *  **SVM-SMOTE (2009, 120+ citations) \[[Paper][Paper 26]\]\[[Code][Code 18]\]** \- SMOTE based on Support Vectors of SVM
 *  **MDO (2015, 150+ citations) \[[Paper][Paper 27]\]\[[Code][Code 19]\]** \- Mahalanobis Distance-based Over-sampling for *Multi-Class* imbalanced problems.

> **NOTE:** See more over-sampling methods at [smote-variants][].

欠采样 | *Under-sampling*

*  **RUS \[[Code][Code 20]\]** \- Random Under-sampling
 *  **CNN (1968, 2100+ citations) \[[Paper][Paper 28]\]\[[Code][Code 21]\]** \- Condensed Nearest Neighbor
 *  **ENN (1972, 1500+ citations) \[[Paper][Paper 29]\] \[[Code][Code 22]\]** \- Edited Condensed Nearest Neighbor
 *  **TomekLink (1976, 870+ citations) \[[Paper][Paper 30]\]\[[Code][Code 23]\]** \- Tomek's modification of Condensed Nearest Neighbor
 *  **NCR (2001, 500+ citations) \[[Paper][Paper 31]\]\[[Code][Code 24]\]** \- Neighborhood Cleaning Rule
 *  **NearMiss-1 & 2 & 3 (2003, 420+ citations) \[[Paper][Paper 32]\]\[[Code][Code 25]\]** \- Several kNN approaches to unbalanced data distributions.
 *  **CNN with TomekLink (2004, 2000+ citations) \[[Paper][Paper 33]\]\[[Code(Java)][Code_Java 2]\]** \- Condensed Nearest Neighbor + TomekLink
 *  **OSS (2007, 2100+ citations) \[[Paper][Paper 34]\]\[[Code][Code 26]\]** \- One Side Selection
 *  **EUS (2009, 290+ citations) \[[Paper][Paper 35]\]** \- Evolutionary Under-sampling
 *  **IHT (2014, 130+ citations) \[[Paper][Paper 36]\]\[[Code][Code 27]\]** \- Instance Hardness Threshold

混合采样 | *Hybrid-sampling*

*  **A Study of the Behavior of Several Methods for Balancing Training Data (2004, 2000+ citations) \[[Paper][Paper 37]\]**
    
    > **NOTE:** extensive experimental evaluation involving 10 different over/under-sampling methods.
    
     *  **SMOTE-Tomek \[[Code][Code 28]\]**
     *  **SMOTE-ENN \[[Code][Code 29]\]**
 *  **SMOTE-RSB (2012, 210+ citations) \[[Paper][Paper 38]\]\[[Code][Code 30]\]** \- Hybrid Preprocessing using SMOTE and Rough Sets Theory
 *  **SMOTE-IPF (2015, 180+ citations) \[[Paper][Paper 39]\]\[[Code][Code 31]\]** \- SMOTE with Iterative-Partitioning Filter

### 代价敏感学习 | Cost-sensitive Learning ###

*  **CSC4.5 (2002, 420+ citations) \[[Paper][Paper 40]\]\[[Code(Java)][Code_Java 3]\]** \- An instance-weighting method to induce cost-sensitive trees
 *  **CSSVM (2008, 710+ citations) \[[Paper][Paper 41]\]\[[Code(Java)][Code_Java 4]\]** \- Cost-sensitive SVMs for highly imbalanced classification
 *  **CSNN (2005, 950+ citations) \[[Paper][Paper 42]\]\[[Code(Java)][Code_Java 5]\]** \- Training cost-sensitive neural networks with methods addressing the class imbalance problem.

### 深度学习 | Deep Learning ###

综述 | *Surveys*

*  A systematic study of the class imbalance problem in convolutional neural networks (2018, 330+ citations) \[[Paper][Paper 43]\]
 *  Survey on deep learning with class imbalance (2019, 50+ citations) \[[Paper][Paper 44]\]
    
    > **NOTE:** a recent comprehensive survey of the class imbalance problem in deep learning.

图数据挖掘 | *Graph Neural Networks*

*  GraphSMOTE: Imbalanced Node Classification on Graphs with Graph Neural Networks (WSDM 2021) \[[Paper][Paper 45]\]\[[Code][Code 32]\]
 *  Topology-Imbalance Learning for Semi-Supervised Node Classification (NeurIPS 2021) \[[Paper][Paper 46]\]\[[Code][Code 33]\]
 *  GraphENS: Neighbor-Aware Ego Network Synthesis for Class-Imbalanced Node Classification (ICLR 2022) \[[Paper][Paper 47]\]\[[Code][Code 34]\]
 *  LTE4G: Long-Tail Experts for Graph Neural Networks (CIKM 2022) \[[Paper][Paper 48]\]\[[Code][Code 35]\]

难例挖掘 | *Hard example mining*

*  Training region-based object detectors with online hard example mining (CVPR 2016, 840+ citations) \[[Paper][Paper 49]\]\[[Code][Code 36]\] - In the later phase of NN training, only do gradient back-propagation for "hard examples" (i.e., with large loss value)

损失函数设计 | *Loss function engineering*

*  Focal loss for dense object detection (ICCV 2017, 2600+ citations) \[[Paper][Paper 50]\]\[[Code (detectron2)][Code _detectron2]\]\[[Code (unofficial)][Code _unofficial]\] - 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.
    
    > **NOTE:** elegant solution, high influence.
 *  Training deep neural networks on imbalanced data sets (IJCNN 2016, 110+ citations) \[[Paper][Paper 51]\] - Mean (square) false error that can equally capture classification errors from both the majority class and the minority class.
 *  Deep imbalanced attribute classification using visual attention aggregation (ECCV 2018, 30+ citation) \[[Paper][Paper 52]\]\[[Code][Code 37]\]
 *  Imbalanced deep learning by minority class incremental rectification (TPAMI 2018, 60+ citations) \[[Paper][Paper 53]\] - 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.
 *  Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss (NIPS 2019, 10+ citations) \[[Paper][Paper 54]\]\[[Code][Code 38]\] - A theoretically-principled label-distribution-aware margin (LDAM) loss motivated by minimizing a margin-based generalization bound.
 *  Gradient harmonized single-stage detector (AAAI 2019, 40+ citations) \[[Paper][Paper 55]\]\[[Code][Code 39]\] - 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.
 *  Class-Balanced Loss Based on Effective Number of Samples (CVPR 2019, 70+ citations) \[[Paper][Paper 56]\]\[[Code][Code 40]\] - a simple and generic class-reweighting mechanism based on Effective Number of Samples.
 *  Influence-Balanced Loss for Imbalanced Visual Classification (ICCV 2021) \[[Paper][Paper 57]\]\[[Code][Code 41]\]
 *  AutoBalance: Optimized Loss Functions for Imbalanced Data (NeurIPS 2021) \[[Paper][Paper 58]\]
 *  Label-Imbalanced and Group-Sensitive Classification under Overparameterization (NeurIPS 2021) \[[Paper][Paper 59]\]\[[Code][Code 42]\]

元学习 | *Meta-learning*

*  Learning to model the tail (NIPS 2017, 70+ citations) \[[Paper][Paper 60]\] - Transfer meta-knowledge from the data-rich classes in the head of the distribution to the data-poor classes in the tail.
 *  Learning to reweight examples for robust deep learning (ICML 2018, 150+ citations) \[[Paper][Paper 61]\]\[[Code][Code 43]\] - Implicitly learn a weight function to reweight the samples in gradient updates of DNN.
    
    > **NOTE:** representative work to solve the class imbalance problem through meta-learning.
 *  Meta-weight-net: Learning an explicit mapping for sample weighting (NIPS 2019) \[[Paper][Paper 62]\]\[[Code][Code 44]\] - Explicitly learn a weight function (with an MLP as the function approximator) to reweight the samples in gradient updates of DNN.
 *  Learning Data Manipulation for Augmentation and Weighting (NIPS 2019) \[[Paper][Paper 63]\]\[[Code][Code 45]\]
 *  Learning to Balance: Bayesian Meta-Learning for Imbalanced and Out-of-distribution Tasks (ICLR 2020) \[[Paper][Paper 64]\]\[[Code][Code 46]\]
 *  MESA: Boost Ensemble Imbalanced Learning with MEta-SAmpler (NeurIPS 2020) \[[Paper][Paper 7]\]\[[Code][Code 1]\]\[[Video][]\]
    
    > **NOTE:** meta-learning-powered ensemble learning

表示学习 | *Representation Learning*

*  Learning deep representation for imbalanced classification (CVPR 2016, 220+ citations) \[[Paper][Paper 65]\]
 *  Supervised Class Distribution Learning for GANs-Based Imbalanced Classification (ICDM 2019) \[[Paper][Paper 66]\]
 *  Decoupling Representation and Classifier for Long-tailed Recognition (ICLR 2020) \[[Paper][Paper 67]\]\[[Code][Code 47]\]
    
    > **NOTE:** interesting findings on representation learning and classifier learning
 *  Supercharging Imbalanced Data Learning With Energy-based Contrastive Representation Transfer (NeurIPS 2021) \[[Paper][Paper 68]\]

后验概率校准 | *Posterior Recalibration*

*  Posterior Re-calibration for Imbalanced Datasets (NeurIPS 2020) \[[Paper][Paper 69]\]\[[Code][Code 48]\]
 *  Long-tail learning via logit adjustment (ICLR 2021) \[[Paper][Paper 70]\]\[[Code][Code 49]\]

半监督/自监督学习 | *Semi/Self-supervised Learning*

*  Rethinking the Value of Labels for Improving Class-Imbalanced Learning (NeurIPS 2020) \[[Paper][Paper 71]\]\[[Code][Code 50]\]\[[Video][Video 1]\]
    
    > **NOTE:** semi-supervised training / self-supervised pre-training helps imbalance learning
 *  Distribution Aligning Refinery of Pseudo-label for Imbalanced Semi-supervised Learning (NeurIPS 2020) \[[Paper][Paper 72]\]\[[Code][Code 51]\]
 *  ABC: Auxiliary Balanced Classifier for Class-imbalanced Semi-supervised Learning (NeurIPS 2021) \[[Paper][Paper 73]\]\[[Code][Code 52]\]
 *  Improving Contrastive Learning on Imbalanced Data via Open-World Sampling (NeurIPS 2021) \[[Paper][Paper 74]\]
 *  DASO: Distribution-Aware Semantics-Oriented Pseudo-label for Imbalanced Semi-Supervised Learning (CVPR 2022) \[[Paper][Paper 75]\]\[[Code][Code 53]\]

课程学习 | *Curriculum Learning*

*  Dynamic Curriculum Learning for Imbalanced Data Classification (ICCV 2019) \[[Paper][Paper 76]\]

双阶段训练 | *Two-phase Training*

*  Brain tumor segmentation with deep neural networks (2017, 1200+ citations) \[[Paper][Paper 77]\]\[[Code (unofficial)][Code _unofficial 1]\]
    
    > Pre-training on balanced dataset, fine-tuning the last output layer before softmax on the original, imbalanced data.

网络结构 | *Network Architecture*

*  BBN: Bilateral-Branch Network with Cumulative Learning for Long-Tailed Visual Recognition (CVPR 2020) \[[Paper][Paper 78]\]\[[Code][Code 54]\]
 *  Class-Imbalanced Deep Learning via a Class-Balanced Ensemble (TNNLS 2021) \[[Paper][Paper 79]\]

深度生成网络 | *Deep Generative Model*

*  Deep Generative Model for Robust Imbalance Classification (CVPR 2020) \[[Paper][Paper 80]\]

不平衡回归 | *Imbalanced Regression*

*  Delving into Deep Imbalanced Regression (ICML 2021) \[[Paper][Paper 81]\]\[[Code][Code 55]\]\[[Video][Video 2]\]
 *  Density-based weighting for imbalanced regression (Machine Learning \[J\], 2021) \[[Paper][Paper 82]\]\[[Code][Code 56]\]

### 异常检测 | Anomaly Detection ###

*  综述 | **Surveys**
    
     *  Anomaly detection: A survey (ACM computing surveys, 2009, 9000+ citations) \[[Paper][Paper 83]\]
     *  A survey of network anomaly detection techniques (2017, 700+ citations) \[[Paper][Paper 84]\]
 *  基于分类的方法 | **Classification-based**
    
     *  One-class SVMs for document classification (JMLR, 2001, 1300+ citations) \[[Paper][Paper 85]\]
     *  One-class Collaborative Filtering (ICDM 2008, 1000+ citations) \[[Paper][Paper 86]\]
     *  Isolation Forest (ICDM 2008, 1000+ citations) \[[Paper][Paper 87]\]
     *  Anomaly Detection using One-Class Neural Networks (2018, 200+ citations) \[[Paper][Paper 88]\]
     *  Anomaly Detection with Robust Deep Autoencoders (KDD 2017, 170+ citations) \[[Paper][Paper 89]\]

## 杂项 | Miscellaneous ##

### 数据集 | Datasets ###

*  **`imbalanced-learn` datasets**
    
    > This collection of datasets is from [imblearn.datasets.fetch\_datasets][imblearn.datasets.fetch_datasets].
    
    <table> 
     <thead> 
      <tr> 
       <th>ID</th> 
       <th>Name</th> 
       <th>Repository &amp; Target</th> 
       <th>Ratio</th> 
       <th>#S</th> 
       <th>#F</th> 
      </tr> 
     </thead> 
     <tbody> 
      <tr> 
       <td>1</td> 
       <td>ecoli</td> 
       <td>UCI, target: imU</td> 
       <td>8.6:1</td> 
       <td>336</td> 
       <td>7</td> 
      </tr> 
      <tr> 
       <td>2</td> 
       <td>optical_digits</td> 
       <td>UCI, target: 8</td> 
       <td>9.1:1</td> 
       <td>5,620</td> 
       <td>64</td> 
      </tr> 
      <tr> 
       <td>3</td> 
       <td>satimage</td> 
       <td>UCI, target: 4</td> 
       <td>9.3:1</td> 
       <td>6,435</td> 
       <td>36</td> 
      </tr> 
      <tr> 
       <td>4</td> 
       <td>pen_digits</td> 
       <td>UCI, target: 5</td> 
       <td>9.4:1</td> 
       <td>10,992</td> 
       <td>16</td> 
      </tr> 
      <tr> 
       <td>5</td> 
       <td>abalone</td> 
       <td>UCI, target: 7</td> 
       <td>9.7:1</td> 
       <td>4,177</td> 
       <td>10</td> 
      </tr> 
      <tr> 
       <td>6</td> 
       <td>sick_euthyroid</td> 
       <td>UCI, target: sick euthyroid</td> 
       <td>9.8:1</td> 
       <td>3,163</td> 
       <td>42</td> 
      </tr> 
      <tr> 
       <td>7</td> 
       <td>spectrometer</td> 
       <td>UCI, target: &gt; =44</td> 
       <td>11:1</td> 
       <td>531</td> 
       <td>93</td> 
      </tr> 
      <tr> 
       <td>8</td> 
       <td>car_eval_34</td> 
       <td>UCI, target: good, v good</td> 
       <td>12:1</td> 
       <td>1,728</td> 
       <td>21</td> 
      </tr> 
      <tr> 
       <td>9</td> 
       <td>isolet</td> 
       <td>UCI, target: A, B</td> 
       <td>12:1</td> 
       <td>7,797</td> 
       <td>617</td> 
      </tr> 
      <tr> 
       <td>10</td> 
       <td>us_crime</td> 
       <td>UCI, target: &gt;0.65</td> 
       <td>12:1</td> 
       <td>1,994</td> 
       <td>100</td> 
      </tr> 
      <tr> 
       <td>11</td> 
       <td>yeast_ml8</td> 
       <td>LIBSVM, target: 8</td> 
       <td>13:1</td> 
       <td>2,417</td> 
       <td>103</td> 
      </tr> 
      <tr> 
       <td>12</td> 
       <td>scene</td> 
       <td>LIBSVM, target: &gt;one label</td> 
       <td>13:1</td> 
       <td>2,407</td> 
       <td>294</td> 
      </tr> 
      <tr> 
       <td>13</td> 
       <td>libras_move</td> 
       <td>UCI, target: 1</td> 
       <td>14:1</td> 
       <td>360</td> 
       <td>90</td> 
      </tr> 
      <tr> 
       <td>14</td> 
       <td>thyroid_sick</td> 
       <td>UCI, target: sick</td> 
       <td>15:1</td> 
       <td>3,772</td> 
       <td>52</td> 
      </tr> 
      <tr> 
       <td>15</td> 
       <td>coil_2000</td> 
       <td>KDD, CoIL, target: minority</td> 
       <td>16:1</td> 
       <td>9,822</td> 
       <td>85</td> 
      </tr> 
      <tr> 
       <td>16</td> 
       <td>arrhythmia</td> 
       <td>UCI, target: 06</td> 
       <td>17:1</td> 
       <td>452</td> 
       <td>278</td> 
      </tr> 
      <tr> 
       <td>17</td> 
       <td>solar_flare_m0</td> 
       <td>UCI, target: M-&gt;0</td> 
       <td>19:1</td> 
       <td>1,389</td> 
       <td>32</td> 
      </tr> 
      <tr> 
       <td>18</td> 
       <td>oil</td> 
       <td>UCI, target: minority</td> 
       <td>22:1</td> 
       <td>937</td> 
       <td>49</td> 
      </tr> 
      <tr> 
       <td>19</td> 
       <td>car_eval_4</td> 
       <td>UCI, target: vgood</td> 
       <td>26:1</td> 
       <td>1,728</td> 
       <td>21</td> 
      </tr> 
      <tr> 
       <td>20</td> 
       <td>wine_quality</td> 
       <td>UCI, wine, target: &lt;=4</td> 
       <td>26:1</td> 
       <td>4,898</td> 
       <td>11</td> 
      </tr> 
      <tr> 
       <td>21</td> 
       <td>letter_img</td> 
       <td>UCI, target: Z</td> 
       <td>26:1</td> 
       <td>20,000</td> 
       <td>16</td> 
      </tr> 
      <tr> 
       <td>22</td> 
       <td>yeast_me2</td> 
       <td>UCI, target: ME2</td> 
       <td>28:1</td> 
       <td>1,484</td> 
       <td>8</td> 
      </tr> 
      <tr> 
       <td>23</td> 
       <td>webpage</td> 
       <td>LIBSVM, w7a, target: minority</td> 
       <td>33:1</td> 
       <td>34,780</td> 
       <td>300</td> 
      </tr> 
      <tr> 
       <td>24</td> 
       <td>ozone_level</td> 
       <td>UCI, ozone, data</td> 
       <td>34:1</td> 
       <td>2,536</td> 
       <td>72</td> 
      </tr> 
      <tr> 
       <td>25</td> 
       <td>mammography</td> 
       <td>UCI, target: minority</td> 
       <td>42:1</td> 
       <td>11,183</td> 
       <td>6</td> 
      </tr> 
      <tr> 
       <td>26</td> 
       <td>protein_homo</td> 
       <td>KDD CUP 2004, minority</td> 
       <td>111:1</td> 
       <td>145,751</td> 
       <td>74</td> 
      </tr> 
      <tr> 
       <td>27</td> 
       <td>abalone_19</td> 
       <td>UCI, target: 19</td> 
       <td>130:1</td> 
       <td>4,177</td> 
       <td>10</td> 
      </tr> 
     </tbody> 
    </table>
 *  **Imbalanced Databases**
    
    Link: [GitHub - gykovacs/common\_datasets: machine learning databases][GitHub - gykovacs_common_datasets_ machine learning databases]

### Github 项目 | Github Repositories ###

#### 算法实现 & 实用程序 & 教程 | *Algorithms & Utilities & Jupyter Notebooks* ####

*  [imbalanced-algorithms][] \- Python-based implementations of algorithms for learning on imbalanced data.
 *  [imbalanced-dataset-sampler][] \- A (PyTorch) imbalanced dataset sampler for oversampling low frequent classes and undersampling high frequent ones.
 *  [class\_imbalance][class_imbalance] \- Jupyter Notebook presentation for class imbalance in binary classification.
 *  [Multi-class-with-imbalanced-dataset-classification][] \- Perform multi-class classification on imbalanced 20-news-group dataset.
 *  [Advanced Machine Learning with scikit-learn: Imbalanced classification and text data][Advanced Machine Learning with scikit-learn_ Imbalanced classification and text data] \- Different approaches to feature selection, and resampling methods for imbalanced data.

#### 论文列表 | *Paper list* ####

*  [Anomaly Detection Learning Resources][] by [yzhao062][] \- Anomaly detection related books, papers, videos, and toolboxes.
 *  [Paper-list-on-Imbalanced-Time-series-Classification-with-Deep-Learning][] \- Imbalanced Time-series Classification

#### 幻灯片 | *Slides* ####

*  [acm\_imbalanced\_learning][acm_imbalanced_learning] \- slides and code for the ACM Imbalanced Learning talk on 27th April 2016 in Austin, TX.

##  ##

[awesome-imbalanced-learning/README\_CN.md at master · ZhiningLiu1998/awesome-imbalanced-learning · GitHub][_ Loss function engineering]

[Link 1]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E7%9B%AE%E5%BD%95
[_ Frameworks and Libraries]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E6%A1%86%E6%9E%B6%E4%B8%8E%E5%BA%93--frameworks-and-libraries
[Python]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#python
[R]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#r
[Java]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#java
[Scalar]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#scalar
[Julia]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#julia
[_ Research Papers]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E7%A0%94%E7%A9%B6%E8%AE%BA%E6%96%87--research-papers
[_ Surveys]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E7%BB%BC%E8%BF%B0--surveys
[_ Ensemble Learning]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E9%9B%86%E6%88%90%E5%AD%A6%E4%B9%A0--ensemble-learning
[_ General ensemble]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E9%80%9A%E7%94%A8%E9%9B%86%E6%88%90%E6%A1%86%E6%9E%B6--general-ensemble
[Boosting _ _ Boosting-based]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E5%9F%BA%E4%BA%8E-boosting-%E7%9A%84%E6%96%B9%E6%B3%95--boosting-based
[Bagging _ _ Bagging-based]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E5%9F%BA%E4%BA%8E-bagging-%E7%9A%84%E6%96%B9%E6%B3%95--bagging-based
[_ Cost-sensitive ensemble]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E5%9F%BA%E4%BA%8E%E4%BB%A3%E4%BB%B7%E6%95%8F%E6%84%9F%E5%AD%A6%E4%B9%A0%E7%9A%84%E6%96%B9%E6%B3%95--cost-sensitive-ensemble
[_ Data resampling]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E6%95%B0%E6%8D%AE%E9%87%8D%E9%87%87%E6%A0%B7--data-resampling
[_ Over-sampling]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E8%BF%87%E9%87%87%E6%A0%B7--over-sampling
[_ Under-sampling]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E6%AC%A0%E9%87%87%E6%A0%B7--under-sampling
[_ Hybrid-sampling]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E6%B7%B7%E5%90%88%E9%87%87%E6%A0%B7--hybrid-sampling
[_ Cost-sensitive Learning]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E4%BB%A3%E4%BB%B7%E6%95%8F%E6%84%9F%E5%AD%A6%E4%B9%A0--cost-sensitive-learning
[_ Deep Learning]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0--deep-learning
[_ Surveys 1]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E7%BB%BC%E8%BF%B0--surveys-1
[_ Graph Neural Networks]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E5%9B%BE%E6%95%B0%E6%8D%AE%E6%8C%96%E6%8E%98--graph-neural-networks
[_ Hard example mining]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E9%9A%BE%E4%BE%8B%E6%8C%96%E6%8E%98--hard-example-mining
[_ Loss function engineering]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E6%8D%9F%E5%A4%B1%E5%87%BD%E6%95%B0%E8%AE%BE%E8%AE%A1--loss-function-engineering
[_ Meta-learning]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E5%85%83%E5%AD%A6%E4%B9%A0--meta-learning
[_ Representation Learning]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E8%A1%A8%E7%A4%BA%E5%AD%A6%E4%B9%A0--representation-learning
[_ Posterior Recalibration]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E5%90%8E%E9%AA%8C%E6%A6%82%E7%8E%87%E6%A0%A1%E5%87%86--posterior-recalibration
[_ Semi_Self-supervised Learning]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E5%8D%8A%E7%9B%91%E7%9D%A3%E8%87%AA%E7%9B%91%E7%9D%A3%E5%AD%A6%E4%B9%A0--semiself-supervised-learning
[_ Curriculum Learning]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E8%AF%BE%E7%A8%8B%E5%AD%A6%E4%B9%A0--curriculum-learning
[_ Two-phase Training]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E5%8F%8C%E9%98%B6%E6%AE%B5%E8%AE%AD%E7%BB%83--two-phase-training
[_ Network Architecture]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E7%BD%91%E7%BB%9C%E7%BB%93%E6%9E%84--network-architecture
[_ Deep Generative Model]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E6%B7%B1%E5%BA%A6%E7%94%9F%E6%88%90%E7%BD%91%E7%BB%9C--deep-generative-model
[_ Imbalanced Regression]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E4%B8%8D%E5%B9%B3%E8%A1%A1%E5%9B%9E%E5%BD%92--imbalanced-regression
[_ Anomaly Detection]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E5%BC%82%E5%B8%B8%E6%A3%80%E6%B5%8B--anomaly-detection
[_ Miscellaneous]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E6%9D%82%E9%A1%B9--miscellaneous
[_ Datasets]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E6%95%B0%E6%8D%AE%E9%9B%86--datasets
[Github _ _ Github Repositories]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#github-%E9%A1%B9%E7%9B%AE--github-repositories
[_ _ _ _ _ Algorithms _ Utilities _ Jupyter Notebooks]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E7%AE%97%E6%B3%95%E5%AE%9E%E7%8E%B0--%E5%AE%9E%E7%94%A8%E7%A8%8B%E5%BA%8F--%E6%95%99%E7%A8%8B--algorithms--utilities--jupyter-notebooks
[_ Paper list]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E8%AE%BA%E6%96%87%E5%88%97%E8%A1%A8--paper-list
[_ Slides]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#%E5%B9%BB%E7%81%AF%E7%89%87--slides
[Contributors]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/README_CN.md#contributors-
[imbalanced-ensemble]: https://imbalanced-ensemble.readthedocs.io/
[Github]: https://github.com/ZhiningLiu1998/imbalanced-ensemble
[Gallery]: https://imbalanced-ensemble.readthedocs.io/en/latest/auto_examples/index.html#
[Paper]: https://arxiv.org/pdf/2111.12776.pdf
[joblib]: https://github.com/joblib/joblib
[scikit-learn]: https://scikit-learn.org/stable/
[imbalanced-learn]: https://imbalanced-learn.org/stable/
[Github 1]: https://github.com/scikit-learn-contrib/imbalanced-learn
[Paper 1]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/www.jmlr.org/papers/volume18/16-365/16-365.pdf
[scikit-learn-contrib]: https://github.com/scikit-learn-contrib
[API reference]: https://imbalanced-learn.org/stable/references/index.html#api
[smote_variants]: https://smote-variants.readthedocs.io/en/latest/
[Github 2]: https://github.com/analyticalmindsltd/smote_variants
[caret]: https://cran.r-project.org/web/packages/caret/index.html
[Documentation]: http://topepo.github.io/caret/index.html
[Github 3]: https://github.com/topepo/caret
[ROSE]: https://cran.r-project.org/web/packages/ROSE/index.html
[Documentation 1]: https://www.rdocumentation.org/packages/ROSE/versions/0.0-3
[ROSE 1]: https://journal.r-project.org/archive/2014-1/menardi-lunardon-torelli.pdf
[DMwR]: https://cran.r-project.org/web/packages/DMwR/index.html
[Documentation 2]: https://www.rdocumentation.org/packages/DMwR/versions/0.4.1
[SMOTE]: https://arxiv.org/pdf/1106.1813.pdf
[KEEL]: https://sci2s.ugr.es/keel/description.php
[Github 4]: https://github.com/SCI2SUGR/KEEL
[Paper 2]: https://sci2s.ugr.es/sites/default/files/ficherosPublicaciones/0758_Alcalaetal-SoftComputing-Keel1.0.pdf
[undersampling]: https://github.com/NestorRV/undersampling
[Documentation 3]: https://nestorrv.github.io/
[Paper 3]: https://www.sci-hub.shop/10.1109/tkde.2008.239
[Paper 4]: https://www.researchgate.net/publication/301596547_Learning_from_imbalanced_data_Open_challenges_and_future_directions
[Paper 5]: https://www.researchgate.net/publication/311977198_Learning_from_class-imbalanced_data_Review_of_methods_and_applications
[Paper 6]: https://arxiv.org/pdf/1909.03500v3.pdf
[Code]: https://github.com/ZhiningLiu1998/self-paced-ensemble
[Slides]: https://zhiningliu.com/files/ICDE_2020_self_paced_ensemble_slides.pdf
[Zhihu]: https://zhuanlan.zhihu.com/p/86891438
[PyPI]: https://pypi.org/project/self-paced-ensemble/
[Paper 7]: https://arxiv.org/pdf/2010.08830.pdf
[Code 1]: https://github.com/ZhiningLiu1998/mesa
[Video]: https://studio.slideslive.com/web_recorder/share/20201020T134559Z__NeurIPS_posters__17343__mesa-effective-ensemble-imbal?s=d3745afc-cfcf-4d60-9f34-63d3d811b55f
[Zhihu 1]: https://zhuanlan.zhihu.com/p/268539195
[Paper 8]: https://sci2s.ugr.es/keel/pdf/algorithm/articulo/2009-IEEE%20TSMCpartB%20Exploratory%20Undersampling%20for%20Class%20Imbalance%20Learning.pdf
[Code 2]: https://github.com/ZhiningLiu1998/imbalanced-ensemble/blob/main/imbalanced_ensemble/ensemble/under_sampling/easy_ensemble.py
[Code 3]: https://github.com/ZhiningLiu1998/imbalanced-ensemble/blob/main/imbalanced_ensemble/ensemble/under_sampling/balance_cascade.py
[Paper 9]: https://sci2s.ugr.es/keel/pdf/algorithm/articulo/1997-JCSS-Schapire-A%20Decision-Theoretic%20Generalization%20of%20On-Line%20Learning%20%28AdaBoost%29.pdf
[Code 4]: https://github.com/scikit-learn/scikit-learn/blob/95d4f0841/sklearn/ensemble/_weight_boosting.py#L285
[Paper 10]: https://sci2s.ugr.es/keel/pdf/algorithm/articulo/2004-SIGKDD-GuoViktor.pdf
[Paper 11]: https://sci2s.ugr.es/keel/pdf/algorithm/congreso/2003-PKDD-SMOTEBoost-ChawlaLazarevicHallBowyer.pdf
[Code 5]: https://github.com/ZhiningLiu1998/imbalanced-ensemble/blob/main/imbalanced_ensemble/ensemble/over_sampling/smote_bagging.py
[Paper 12]: https://sci2s.ugr.es/keel/pdf/algorithm/articulo/2011-IEEE%20TSMC%20partC-%20GalarFdezBarrenecheaBustinceHerrera.pdf
[Paper 13]: https://www.ele.uri.edu/faculty/he/PDFfiles/ramoboost.pdf
[Code 6]: https://github.com/dialnd/imbalanced-algorithms/blob/master/ramo.py#L133
[Paper 14]: https://sci2s.ugr.es/keel/pdf/algorithm/articulo/2010-IEEE%20TSMCpartA-RUSBoost%20A%20Hybrid%20Approach%20to%20Alleviating%20Class%20Imbalance.pdf
[Code 7]: https://github.com/ZhiningLiu1998/imbalanced-ensemble/blob/main/imbalanced_ensemble/ensemble/under_sampling/rus_boost.py
[Paper 15]: https://sci2s.ugr.es/keel/pdf/algorithm/articulo/2012-wang-IEEE_SMC_B.pdf
[Paper 16]: https://sci2s.ugr.es/keel/pdf/algorithm/articulo/2013-galar-PR.pdf
[Paper 17]: https://sci2s.ugr.es/keel/pdf/algorithm/articulo/1996-ML-Breiman-Bagging%20Predictors.pdf
[Code 8]: https://github.com/scikit-learn/scikit-learn/blob/95d4f0841/sklearn/ensemble/_bagging.py#L433
[Paper 18]: https://sci2s.ugr.es/keel/pdf/algorithm/congreso/2009-IEEE%20CIDM-WangYao.pdf
[Code 9]: https://github.com/ZhiningLiu1998/imbalanced-ensemble/blob/main/imbalanced_ensemble/ensemble/under_sampling/under_bagging.py
[Code 10]: https://github.com/ZhiningLiu1998/imbalanced-ensemble/blob/main/imbalanced_ensemble/ensemble/over_sampling/over_bagging.py
[Paper 19]: https://www.researchgate.net/profile/Salvatore-Stolfo/publication/2628569_AdaCost_Misclassification_Cost-sensitive_Boosting/links/0fcfd50ca581d7016f000000/AdaCost-Misclassification-Cost-sensitive-Boosting.pdf
[Code 11]: https://github.com/ZhiningLiu1998/imbalanced-ensemble/blob/main/imbalanced_ensemble/ensemble/reweighting/adacost.py
[Paper 20]: https://proceedings.neurips.cc/paper/1998/file/df12ecd077efc8c23881028604dbb8cc-Paper.pdf
[Code 12]: https://github.com/ZhiningLiu1998/imbalanced-ensemble/blob/main/imbalanced_ensemble/ensemble/reweighting/adauboost.py
[Paper 21]: https://www.researchgate.net/profile/Michael-Jones-66/publication/2539888_Fast_and_Robust_Classification_using_Asymmetric_AdaBoost_and_a_Detector_Cascade/links/540731780cf23d9765a83ec1/Fast-and-Robust-Classification-using-Asymmetric-AdaBoost-and-a-Detector-Cascade.pdf
[Code 13]: https://github.com/ZhiningLiu1998/imbalanced-ensemble/blob/main/imbalanced_ensemble/ensemble/reweighting/asymmetric_boost.py
[Code 14]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/over_sampling/_random_over_sampler.py
[Code 15]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/over_sampling/_smote.py#L36
[Paper 22]: https://sci2s.ugr.es/keel/keel-dataset/pdfs/2005-Han-LNCS.pdf
[Code 16]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/over_sampling/_smote.py#L220
[Paper 23]: https://sci2s.ugr.es/keel/pdf/algorithm/congreso/2008-He-ieee.pdf
[Code 17]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/over_sampling/_adasyn.py
[Paper 24]: https://sci2s.ugr.es/keel/pdf/algorithm/congreso/stefanowski_selective_2008.pdf
[Code_Java]: https://github.com/SCI2SUGR/KEEL/blob/master/src/keel/Algorithms/ImbalancedClassification/Resampling/SPIDER/SPIDER.java#L57
[Paper 25]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/150.214.190.154/keel/keel-dataset/pdfs/2009-Bunkhumpornpat-LNCS.pdf
[Code_Java 1]: https://github.com/SCI2SUGR/KEEL/blob/master/src/keel/Algorithms/ImbalancedClassification/Resampling/Safe_Level_SMOTE/Safe_Level_SMOTE.java#L58
[Paper 26]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/ousar.lib.okayama-u.ac.jp/files/public/1/19617/20160528004522391723/IWCIA2009_A1005.pdf
[Code 18]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/over_sampling/_smote.py#L417
[Paper 27]: https://ieeexplore.ieee.org/abstract/document/7163639
[Code 19]: https://github.com/analyticalmindsltd/smote_variants/blob/dedbc3d00b266954fedac0ae87775e1643bc920a/smote_variants/_smote_variants.py#L14513
[smote-variants]: https://github.com/analyticalmindsltd/smote_variants#references
[Code 20]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_random_under_sampler.py
[Paper 28]: https://pdfs.semanticscholar.org/7c37/71fd6829630cf450af853df728ecd8da4ab2.pdf?_ga=2.137274553.882046879.1583413150-1712662047.1583413150
[Code 21]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_condensed_nearest_neighbour.py
[Paper 29]: https://sci2s.ugr.es/keel/dataset/includes/catImbFiles/1972-Wilson-IEEETSMC.pdf
[Code 22]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_edited_nearest_neighbours.py
[Paper 30]: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4309452
[Code 23]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_tomek_links.py
[Paper 31]: https://sci2s.ugr.es/keel/pdf/algorithm/congreso/2001-Laurikkala-LNCS.pdf
[Code 24]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_neighbourhood_cleaning_rule.py
[Paper 32]: https://sci2s.ugr.es/keel/pdf/specific/congreso/jzhang.pdf
[Code 25]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_nearmiss.py
[Paper 33]: https://storm.cis.fordham.edu/~gweiss/selected-papers/batista-study-balancing-training-data.pdf
[Code_Java 2]: https://github.com/SCI2SUGR/KEEL/blob/master/src/keel/Algorithms/ImbalancedClassification/Resampling/CNN_TomekLinks/CNN_TomekLinks.java#L58
[Paper 34]: https://sci2s.ugr.es/keel/pdf/specific/congreso/kubat97addressing.pdf
[Code 26]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_one_sided_selection.py
[Paper 35]: https://www.mitpressjournals.org/doi/pdfplus/10.1162/evco.2009.17.3.275
[Paper 36]: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.649.8727&rep=rep1&type=pdf
[Code 27]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/under_sampling/_prototype_selection/_instance_hardness_threshold.py
[Paper 37]: https://www.researchgate.net/profile/Ronaldo-Prati/publication/220520041_A_Study_of_the_Behavior_of_Several_Methods_for_Balancing_machine_Learning_Training_Data/links/0d22cd91c989507054a2cf3b/A-Study-of-the-Behavior-of-Several-Methods-for-Balancing-machine-Learning-Training-Data.pdf
[Code 28]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/combine/_smote_tomek.py
[Code 29]: https://github.com/scikit-learn-contrib/imbalanced-learn/blob/master/imblearn/combine/_smote_enn.py
[Paper 38]: https://sci2s.ugr.es/sites/default/files/ficherosPublicaciones/1434_2012-Ramentol-KAIS.pdf
[Code 30]: https://smote-variants.readthedocs.io/en/latest/_modules/smote_variants/_smote_variants.html#SMOTE_RSB
[Paper 39]: https://sci2s.ugr.es/sites/default/files/ficherosPublicaciones/1824_2015-INS-Saez.pdf
[Code 31]: https://smote-variants.readthedocs.io/en/latest/_modules/smote_variants/_smote_variants.html#SMOTE_IPF
[Paper 40]: https://www.sci-hub.shop/10.1109/tkde.2002.1000348
[Code_Java 3]: https://github.com/SCI2SUGR/KEEL/blob/master/src/keel/Algorithms/ImbalancedClassification/CSMethods/C45CS/C45CS.java#L48
[Paper 41]: https://sci2s.ugr.es/keel/pdf/algorithm/articulo/2009-Chawla-IEEE_TSMCB-svm-imbalance.pdf
[Code_Java 4]: https://github.com/SCI2SUGR/KEEL/blob/master/src/keel/Algorithms/ImbalancedClassification/CSMethods/C_SVMCost/svmClassifierCost.java#L60
[Paper 42]: https://sci2s.ugr.es/keel/pdf/algorithm/articulo/2006%20-%20IEEE_TKDE%20-%20Zhou_Liu.pdf
[Code_Java 5]: https://github.com/SCI2SUGR/KEEL/blob/master/src/keel/Algorithms/ImbalancedClassification/CSMethods/MLPerceptronBackpropCS/MLPerceptronBackpropCS.java#L49
[Paper 43]: https://arxiv.org/pdf/1710.05381.pdf
[Paper 44]: https://www.researchgate.net/publication/332165523_Survey_on_deep_learning_with_class_imbalance
[Paper 45]: https://dl.acm.org/doi/pdf/10.1145/3437963.3441720
[Code 32]: https://github.com/TianxiangZhao/GraphSmote
[Paper 46]: https://proceedings.neurips.cc//paper/2021/file/fa7cdfad1a5aaf8370ebeda47a1ff1c3-Paper.pdf
[Code 33]: https://github.com/victorchen96/renode
[Paper 47]: https://openreview.net/pdf?id=MXEl7i-iru
[Code 34]: https://github.com/JoonHyung-Park/GraphENS
[Paper 48]: https://arxiv.org/pdf/2208.10205.pdf
[Code 35]: https://github.com/SukwonYun/LTE4G
[Paper 49]: https://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Shrivastava_Training_Region-Based_Object_CVPR_2016_paper.pdf
[Code 36]: https://github.com/abhi2610/ohemh
[Paper 50]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/openaccess.thecvf.com/content_ICCV_2017/papers/Lin_Focal_Loss_for_ICCV_2017_paper.pdf
[Code _detectron2]: https://github.com/facebookresearch/detectron2
[Code _unofficial]: https://github.com/clcarwin/focal_loss_pytorch
[Paper 51]: https://www.researchgate.net/publication/309778930_Training_deep_neural_networks_on_imbalanced_data_sets
[Paper 52]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/openaccess.thecvf.com/content_ECCV_2018/papers/Nikolaos_Sarafianos_Deep_Imbalanced_Attribute_ECCV_2018_paper.pdf
[Code 37]: https://github.com/cvcode18/imbalanced_learning
[Paper 53]: https://arxiv.org/pdf/1804.10851.pdf
[Paper 54]: https://papers.nips.cc/paper/8435-learning-imbalanced-datasets-with-label-distribution-aware-margin-loss.pdf
[Code 38]: https://github.com/kaidic/LDAM-DRW
[Paper 55]: https://arxiv.org/pdf/1811.05181.pdf
[Code 39]: https://github.com/libuyu/GHM_Detection
[Paper 56]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/openaccess.thecvf.com/content_CVPR_2019/papers/Cui_Class-Balanced_Loss_Based_on_Effective_Number_of_Samples_CVPR_2019_paper.pdf
[Code 40]: https://github.com/richardaecn/class-balanced-loss
[Paper 57]: https://arxiv.org/pdf/2110.02444.pdf
[Code 41]: https://github.com/pseulki/IB-Loss
[Paper 58]: https://openreview.net/pdf?id=ebQXflQre5a
[Paper 59]: https://proceedings.neurips.cc//paper/2021/file/9dfcf16f0adbc5e2a55ef02db36bac7f-Paper.pdf
[Code 42]: https://github.com/orparask/VS-Loss
[Paper 60]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/papers.nips.cc/paper/7278-learning-to-model-the-tail.pdf
[Paper 61]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/proceedings.mlr.press/v80/ren18a/ren18a.pdf
[Code 43]: https://github.com/uber-research/learning-to-reweight-examples
[Paper 62]: https://papers.nips.cc/paper/8467-meta-weight-net-learning-an-explicit-mapping-for-sample-weighting.pdf
[Code 44]: https://github.com/xjtushujun/meta-weight-net
[Paper 63]: https://www.cs.cmu.edu/~zhitingh/data/neurips19_data_manip_preprint.pdf
[Code 45]: https://github.com/tanyuqian/learning-data-manipulation
[Paper 64]: https://openreview.net/attachment?id=rkeZIJBYvr&name=original_pdf
[Code 46]: https://github.com/haebeom-lee/l2b
[Paper 65]: https://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Huang_Learning_Deep_Representation_CVPR_2016_paper.pdf
[Paper 66]: https://ieeexplore.ieee.xilesou.top/abstract/document/8970900
[Paper 67]: https://arxiv.org/pdf/1910.09217.pdf
[Code 47]: https://github.com/facebookresearch/classifier-balancing
[Paper 68]: https://proceedings.neurips.cc//paper/2021/file/b151ce4935a3c2807e1dd9963eda16d8-Paper.pdf
[Paper 69]: https://arxiv.org/pdf/2010.11820.pdf
[Code 48]: https://github.com/GT-RIPL/UNO-IC
[Paper 70]: https://arxiv.org/pdf/2007.07314v1.pdf
[Code 49]: https://github.com/google-research/google-research/tree/master/logit_adjustment
[Paper 71]: https://arxiv.org/pdf/2006.07529.pdf
[Code 50]: https://github.com/YyzHarry/imbalanced-semi-self
[Video 1]: https://www.youtube.com/watch?v=XltXZ3OZvyI&feature=youtu.be
[Paper 72]: https://arxiv.org/pdf/2007.08844.pdf
[Code 51]: https://github.com/bbuing9/DARP
[Paper 73]: https://proceedings.neurips.cc//paper/2021/file/3953630da28e5181cffca1278517e3cf-Paper.pdf
[Code 52]: https://github.com/leehyuck/abc
[Paper 74]: https://proceedings.neurips.cc//paper/2021/file/2f37d10131f2a483a8dd005b3d14b0d9-Paper.pdf
[Paper 75]: https://arxiv.org/pdf/2106.05682
[Code 53]: https://github.com/ytaek-oh/daso
[Paper 76]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/openaccess.thecvf.com/content_ICCV_2019/papers/Wang_Dynamic_Curriculum_Learning_for_Imbalanced_Data_Classification_ICCV_2019_paper.pdf
[Paper 77]: https://arxiv.org/pdf/1505.03540.pdf
[Code _unofficial 1]: https://github.com/naldeborgh7575/brain_segmentation
[Paper 78]: https://arxiv.org/pdf/1912.02413.pdf
[Code 54]: https://github.com/Megvii-Nanjing/BBN
[Paper 79]: https://ieeexplore.ieee.org/abstract/document/9416240
[Paper 80]: https://openaccess.thecvf.com/content_CVPR_2020/papers/Wang_Deep_Generative_Model_for_Robust_Imbalance_Classification_CVPR_2020_paper.pdf
[Paper 81]: https://arxiv.org/pdf/2102.09554.pdf
[Code 55]: https://github.com/YyzHarry/imbalanced-regression
[Video 2]: https://www.youtube.com/watch?v=grJGixofQRU
[Paper 82]: https://link.springer.com/article/10.1007/s10994-021-06023-5
[Code 56]: https://github.com/SteiMi/density-based-weighting-for-imbalanced-regression
[Paper 83]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/cinslab.com/wp-content/uploads/2019/03/xiaorong.pdf
[Paper 84]: https://www.gta.ufrj.br/~alvarenga/files/CPE826/Ahmed2016-Survey.pdf
[Paper 85]: https://github.com/ZhiningLiu1998/awesome-imbalanced-learning/blob/master/www.jmlr.org/papers/volume2/manevitz01a/manevitz01a.pdf
[Paper 86]: https://cseweb.ucsd.edu/classes/fa17/cse291-b/reading/04781145.pdf
[Paper 87]: https://cs.nju.edu.cn/zhouzh/zhouzh.files/publication/icdm08b.pdf?q=isolation-forest
[Paper 88]: https://arxiv.org/pdf/1802.06360.pdf
[Paper 89]: https://pdfs.semanticscholar.org/c112/b06d3dac590b4cc111e5ec9c805d0b086c6e.pdf
[imblearn.datasets.fetch_datasets]: https://imbalanced-learn.org/stable/references/generated/imblearn.datasets.fetch_datasets.html
[GitHub - gykovacs_common_datasets_ machine learning databases]: https://github.com/gykovacs/mldb
[imbalanced-algorithms]: https://github.com/dialnd/imbalanced-algorithms
[imbalanced-dataset-sampler]: https://github.com/ufoym/imbalanced-dataset-sampler
[class_imbalance]: https://github.com/wangz10/class_imbalance
[Multi-class-with-imbalanced-dataset-classification]: https://github.com/javaidnabi31/Multi-class-with-imbalanced-dataset-classification
[Advanced Machine Learning with scikit-learn_ Imbalanced classification and text data]: https://github.com/amueller/ml-workshop-4-of-4
[Anomaly Detection Learning Resources]: https://github.com/yzhao062/anomaly-detection-resources
[yzhao062]: https://github.com/yzhao062
[Paper-list-on-Imbalanced-Time-series-Classification-with-Deep-Learning]: https://github.com/danielgy/Paper-list-on-Imbalanced-Time-series-Classification-with-Deep-Learning
[acm_imbalanced_learning]: https://github.com/timgasser/acm_imbalanced_learning