mlsanityData to 7 Oct 2026

Ensembles of GANs for synthetic training data generation

arXiv 2104.11797 ↗First posted 23 Apr 2021cs.CV cs.LG eess.IVcited by 6

Not on the map. Why?

Abstract (arXiv)

Insufficient training data is a major bottleneck for most deep learning practices, not least in medical imaging where data is difficult to collect and publicly available datasets are scarce due to ethics and privacy. This work investigates the use of synthetic images, created by generative adversarial networks (GANs), as the only source of training data. We demonstrate that for this application, it is of great importance to make use of multiple GANs to improve the diversity of the generated data, i.e. to sufficiently cover the data distribution. While a single GAN can generate seemingly diverse image content, training on this data in most cases lead to severe over-fitting. We test the impact of ensembled GANs on synthetic 2D data as well as common image datasets (SVHN and CIFAR-10), and using both DCGANs and progressively growing GANs. As a specific use case, we focus on synthesizing digital pathology patches to provide anonymized training data.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 21 of its 32 references that are in the map. First posted on arXiv 23 Apr 2021.

Builds on

21 of 32 references.

References · 21

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  4. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
  5. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  6. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,048
  7. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,618
  8. 2018Self-Attention Generative Adversarial Networkscited by 809
  9. 2018GAN-based Synthetic Medical Image Augmentation for increased CNN Performance in Liver Lesion Classificationcited by 242
  10. 2018Pros and Cons of GAN Evaluation Measurescited by 224
  11. 2017VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learningcited by 199
  12. 2018Medical Image Synthesis for Data Augmentation and Anonymization using Generative Adversarial Networkscited by 146
9 more
  1. 2018GAN Augmentation: Augmenting Training Data using Generative Adversarial Networkscited by 107
  2. 2017PacGAN: The power of two samples in generative adversarial networkscited by 104
  3. 2018How good is my GAN?cited by 74
  4. 2020Diverse Image Generation via Self-Conditioned GANscited by 54
  5. 2017AdaGAN: Boosting Generative Modelscited by 34
  6. 2017Synthetic Medical Images from Dual Generative Adversarial Networkscited by 32
  7. 2017Boosted Generative Modelscited by 18
  8. 2018Generating Artificial Data for Private Deep Learningcited by 9
  9. 2016Ensembles of Generative Adversarial Networkscited by 8

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 2; 2023: 2; 2024: 0; 2025: 1.

6 papers.

  1. 2023Beware of diffusion models for synthesizing medical images -- A comparison with GANs in terms of memorizing brain MRI and chest x-ray imagescited by 23
  2. 2023Brain tumor segmentation using synthetic MR images -- A comparison of GANs and diffusion modelscited by 19
  3. 2022Rethinking Surgical Instrument Segmentation: A Background Image Can Be All You Needcited by 16
  4. 2021Synthetic Data for Model Selectioncited by 6
  5. 2022Does an ensemble of GANs lead to better performance when training segmentation networks with synthetic images?cited by 2
  6. 2025Beyond a Single Mode: GAN Ensembles for Diverse Medical Data Generationcited by 1