Effective data generation for imbalanced learning using conditional generative adversarial networks

By Georgios Douzas, Fernando Bacao. 2018

576 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Media Technology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Eugene Bagdasarian
2. Fei Wang
3. Vitaly Shmatikov
4. Chris Dongjoo Kim
5. Dhiraj Neupane
6. Adamu Ali-Gombe
7. Sook Yoon
8. Alvaro Fuentes

In Computer Vision and Pattern Recognition

1. Chris Dongjoo Kim
2. Fei Wang
3. Dhiraj Neupane
4. M. Shafie‐khah
5. Adamu Ali-Gombe
6. Sook Yoon
7. Alvaro Fuentes
8. Su Ruan

In Media Technology

1. Adamu Ali-Gombe
2. Sook Yoon
3. Alvaro Fuentes
4. Su Ruan
5. Stefan Lessmann
6. Fei Wang
7. Vitaly Shmatikov
8. Gunhee Kim

Citations in the corpus

• Differential Privacy Has Disparate Impact on Model Accuracy (2019, background): “Models that generate artificial data points [11] from the existing data are incompatible with DP. Recent research [10, 20] aims to add fairness and DP to post-processing [17] and in-processing [2] algorithms.”
• Imbalanced Continual Learning with Partitioning Reservoir Sampling (2020, background): “More aggressively, there have also been directions that populate the minority samples via generation to avoid overfitting in the minority [14,58,23].”
• A review: Deep learning for medical image segmentation using multi-modality fusion (2019, background): “There are three main approaches: under-sampling the negative class [33]or up-sampling the negative class [17] and SMOTE (Synthetic Minority Over-sampling Technique) [10] generating synthetic samples along the line segment that joins minority class samples.”
• Conditional Wasserstein GAN-based Oversampling of Tabular Data for Imbalanced Learning (2020, methodology): “One model, the generator G , is tasked with GANs have rapidly gained popularity as a method for modelling complex data distributions and have achieved impressive results, especially in the realm of images [29].”
• Bearing Fault Detection and Diagnosis Using Case Western Reserve University Dataset With Deep Learning Approaches: A Review (2020, background): “Though it is quite challenging to deal with the complex or imbalanced data, the imbalanced data are the characteristics of multiple real-world applications such as medical diagnosis, fraud detection, machinery fault detection, pricing catastrophe [39], [40].”
• A Comprehensive Survey of Image Augmentation Techniques for Deep Learning (2022, methodology): “Model-based Unconditional DCGAN [41], [42, 43, 44] Label-conditional BDA [45], ImbCGAN [46], BAGAN [47], DAGAN [48], MFC-GAN [49], IDA-GAN [50].”
• MFC-GAN: Class-imbalanced dataset classification using Multiple Fake Class Generative Adversarial Network (2019, methodology): “This could occur in a binary classification or a multi-classification task [1].”
• Generative adversarial networks and convolutional neural networks based weather classification model for day ahead short-term photovoltaic power forecasting (2019, methodology): “As a simple generative model, Random Oversampling (RO) creates new data via copying random minority class examples [30].”