Corpus ID: 237635520

Evolutionary Generative Adversarial Networks based on New Fitness Function and Generic Crossover Operator

  title={Evolutionary Generative Adversarial Networks based on New Fitness Function and Generic Crossover Operator},
  author={Junjie Li and Jingyao Li and Wenbo Zhou and Shuai L{\"u}},
Evolutionary generative adversarial networks (EGAN) attempts to alleviate mode collapse and vanishing gradient that plague generative adversarial networks by introducing evolutionary computation. However, E-GAN lacks a reasonable evaluation mechanism, which limits its effect. Moreover, E-GAN only contains mutation operators in its evolutionary step, while ignoring crossover operators. The crossover operator generates more competitive individuals by combining the good traits of multiple… Expand


Evolutionary Generative Adversarial Networks with Crossover Based Knowledge Distillation
This paper proposes a general crossover operator, which can be widely applied to GANs using evolutionary strategies, and designs an evolutionary GAN framework named C-GAN based on it, and combines the crossover operator with evolutionary generative adversarial networks (E-GAN) to implement the evolutionaryGenerative Adversarial Networks with crossover (CE-GAN). Expand
Spatial evolutionary generative adversarial networks
A superior evolutionary GANs training method is contributed, Mustangs, that eliminates the single loss function used across Lipizzaner's grid and proposes to combine mutation and population approaches to diversity improvement. Expand
Evolutionary Generative Adversarial Networks
A novel GAN framework called evolutionary GANs (E-GANs) is proposed for stable GAN training and improved generative performance, which overcomes the limitations of an individual adversarial training objective and always preserves the well-performing offspring, contributing to progress in, and the success of GAns. Expand
Neuroevolution of Generative Adversarial Networks
The evolutionary pressure is used to guide the training of GANs to build robust models, leveraging the quality of results, and providing a more stable training, and these proposals can automatically provide useful architectural definitions, avoiding the manual discovery of suitable models for GAns. Expand
Enhanced Evolutionary Generative Adversarial Networks
This work represents the concept of realness as a realness distribution rather than a single scalar so that the corresponding discriminator estimates realness of an input sample from multiple angles, providing more informative guidance to generator. Expand
COEGAN: evaluating the coevolution effect in generative adversarial networks
COEGAN is a model that uses neuroevolution and coevolution in the GAN training algorithm to provide a more stable training method and the automatic design of neural network architectures and is able to discover efficient architectures in the Fashion-MNIST and MNIST datasets. Expand
Exploring the evolution of GANs through quality diversity
This paper proposes the application of a quality-diversity algorithm in the evolution of GANs, based on the Novelty Search with Local Competition (NSLC) algorithm, adapting the concepts used in COEGAN to this new proposal. Expand
EvolGAN: Evolutionary Generative Adversarial Networks
We propose to use a quality estimator and evolutionary methods to search the latent space of generative adversarial networks trained on small, difficult datasets, or both. The new method leads to theExpand
GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
This work proposes a two time-scale update rule (TTUR) for training GANs with stochastic gradient descent on arbitrary GAN loss functions and introduces the "Frechet Inception Distance" (FID) which captures the similarity of generated images to real ones better than the Inception Score. Expand
CatGAN: Category-aware Generative Adversarial Networks with Hierarchical Evolutionary Learning for Category Text Generation
A category-aware GAN (CatGAN) which consists of an efficient category- aware model for category text generation and a hierarchical evolutionary learning algorithm for training the authors' model. Expand