mlsanityData to 7 Oct 2026

Evolving GAN Formulations for Higher Quality Image Synthesis

arXiv 2102.08578 ↗First posted 17 Feb 2021cs.NE cs.CV cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

Generative Adversarial Networks (GANs) have extended deep learning to complex generation and translation tasks across different data modalities. However, GANs are notoriously difficult to train: Mode collapse and other instabilities in the training process often degrade the quality of the generated results, such as images. This paper presents a new technique called TaylorGAN for improving GANs by discovering customized loss functions for each of its two networks. The loss functions are parameterized as Taylor expansions and optimized through multiobjective evolution. On an image-to-image translation benchmark task, this approach qualitatively improves generated image quality and quantitatively improves two independent GAN performance metrics. It therefore forms a promising approach for applying GANs to more challenging tasks in the future.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 27 of its 51 references that are in the map. First posted on arXiv 17 Feb 2021.

Builds on

27 of 51 references.

References · 27

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  4. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
  5. 2013Auto-Encoding Variational Bayescited by 3,980
  6. 2017Improved Training of Wasserstein GANscited by 2,223
  7. 2014Conditional Generative Adversarial Netscited by 2,152
  8. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,048
  9. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,618
  10. 2017High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANscited by 1,176
  11. 2016Least Squares Generative Adversarial Networkscited by 1,000
  12. 2018Regularized Evolution for Image Classifier Architecture Searchcited by 902
15 more
  1. 2016Conditional Image Generation with PixelCNN Decoderscited by 859
  2. 2016InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Netscited by 858
  3. 2016Generative Adversarial Text to Image Synthesiscited by 735
  4. 2015Autoencoding beyond pixels using a learned similarity metriccited by 407
  5. 2018Pros and Cons of GAN Evaluation Measurescited by 224
  6. 2016Unrolled Generative Adversarial Networkscited by 208
  7. 2017Evolving Deep Neural Networkscited by 93
  8. 2018Evolving Mario Levels in the Latent Space of a Deep Convolutional Generative Adversarial Networkcited by 76
  9. 2017Many Paths to Equilibrium: GANs Do Not Need to Decrease a Divergence At Every Stepcited by 45
  10. 2019Improved Training Speed, Accuracy, and Data Utilization Through Loss Function Optimizationcited by 28
  11. 2019Evolutionary Neural AutoML for Deep Learningcited by 20
  12. 2017On the Effectiveness of Least Squares Generative Adversarial Networkscited by 18
  13. 2018Learning to Repair Software Vulnerabilities with Generative Adversarial Networkscited by 11
  14. 2018Enhanced Optimization with Composite Objectives and Novelty Selectioncited by 3
  15. 2020Optimizing Loss Functions Through Multivariate Taylor Polynomial Parameterizationcited by 3

Built on by

Cited by year

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

3 papers.

  1. 2022Evolving SimGANs to Improve Abnormal Electrocardiogram Classificationcited by 1
  2. 2024GANetic Loss for Generative Adversarial Networks with a Focus on Medical Applicationscited by 1
  3. 2024Towards evolution of Deep Neural Networks through contrastive Self-Supervised learningcited by 0