CDE-GAN: Cooperative Dual Evolution-Based Generative Adversarial Network

Abstract

Generative adversarial networks (GANs) have been a popular deep generative model for real-word applications. Despite many recent efforts on GANs have been contributed, however, mode collapse and instability of GANs are still open problems caused by their adversarial optimization difficulties. In this paper, motivated by the cooperative co-evolutionary algorithm, we propose a Cooperative Dual Evolution based Generative Adversarial Network (CDE-GAN) to circumvent these drawbacks. In essence, CDE-GAN incorporates dual evolution with respect to generator(s) and discriminators into a unified evolutionary adversarial framework, thus it exploits the complementary properties and injects dual mutation diversity into training to steadily diversify the estimated density in capturing multi-modes, and to improve generative performance. Specifically, CDE-GAN decomposes the complex adversarial optimization problem into two subproblems (generation and discrimination), and each subproblem is solved with a separated subpopulation (E-Generators and E-Discriminators), evolved by an individual evolutionary algorithm. Additionally, to keep the balance between E-Generators and E-Discriminators, we proposed a Soft Mechanism to cooperate them to conduct effective adversarial training. Extensive experiments on one synthetic dataset and three real-world benchmark image datasets, demonstrate that the proposed CDE-GAN achieves the competitive and superior performance in generating good quality and diverse samples over baselines. The code and more generated results are available at our project homepage https: //shiming-chen.github.io/CDE-GAN-website/CDE-GAN.html.

Spoofify: A Hybrid Anti-Spoofing Facial Recognition System

“…Spoofify aims to prevent unauthorized access to facial recognition systems by leveraging the innovative framework of unsupervised GANs [4]. Figure 2 illustrates the proposed architecture of GANs with its generator and discriminator networks.…”

EGANS: Evolutionary Generative Adversarial Network Search for Zero-Shot Learning

“…Co-evolutionary algorithms (coEA) overcome GAN optimization pathologies by evolving two populations: a population of generators and a population of discriminators [48], [50], [53], [62]- [64]. Schmiedlechner et al [48] proposed Lipizzaner by training a two-dimensional grid of GANs with a distributed evolutionary algorithm.…”

“…Based on neuro-evolution and coevolution in the GAN training, Costa et al [50] devised COEGAN to provide a more stable training method and the automatic design of neural network architectures. To enable stable co-evolution between generator and discriminator, Chen et al [53] developed CDE-GAN by conducting adversarial multi-objective optimization. However, the aforementioned CoEA-based GANs only take quality and diversity into consideration.…”

“…Considering the success of evolutionary computation (EC) for the optimization GANs [48]- [53] and NAS [54]- [60], evolutionary neural architecture search (ENAS) based GANs models are introduced to improve the optimization and generalization of GANs recently [42], [61]. Although these methods can synthesize realistic and sharp images, they are not yet able to generate images of sufficient quality to train classifiers [8], [14].…”

“…Furthermore, existing NAS-based GANs are limited by the specified adversarial optimization strategy, resulting in instability and sub-optimal GAN performance. Motivated by the co-evolutionary algorithms that address GAN optimization pathologies effectively by evolving two populations: a population of generators and a population of discriminators [48], [50], [53], [62]- [64], we attempt to design an ENAS algorithm with cooperative dual evolution paradigm to automatically find superior multi-layer perceptron (MLP) architecture-based GAN models, which are effective in stable optimization and generalization for generative ZSL methods.…”