Improving generative adversarial network with multiple generators by evolutionary algorithms
Pages 761 - 769
Abstract
Generative Adversarial Network (GAN) is a novel class of deep generative models that has recently gained significant attention. However, the original GAN with one generator can easily get trapped into the mode collapsing problem, which could cause the generator only to produce similar images. This paper proposed a combination of GAN and an evolutionary algorithm to overcome the mode collapsing problem. In our approach, multiple generator networks are trained with the evolutionary strategy (ES), an evolution algorithm. The discriminator network distinguishes if the image comes from the real dataset or not. An additional classifier network is implemented to distinguish different generators. The mutations in the evolutionary strategy and the additional classifier network keep the diversity among generators. We term our approach the Evolution-GAN. In this paper, we conduct experiments on 2D synthetic data to verify that the Evolution-GAN overcomes the mode collapsing problem. Furthermore, experiments on MNIST datasets are implemented to compare the performance of Evolution-GAN, the original GAN, and Deep Convolutional GAN(DCGAN) and Evolutionary GAN.
References
[1]
Goodfellow I et al. Generative adversarial nets Advances in neural information processing systems 2014 Red Hook NY Curran
[2]
Kossaifi J et al (2018) Gagan: geometry-aware generative adversarial networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 878-887
[3]
Wang X and Gupta A Generative image modeling using style and structure adversarial networks. European conference on computer vision 2016 Cham Springer
[4]
Nguyen-Quynh T-T, Do NT, Soo-Hyung K (2020) MLEU: multi-level embedding U-Net for fully automatic image colorization. In: Proceedings of the 4th International Conference on Machine Learning and Soft Computing
[5]
Chai L, Bau D, Lim S-N, and Isola P What makes fake images detectable? Understanding properties that generalize European conference on computer vision 2020 Berlin Springer 103-120
[6]
Radford A, Luke M, Soumith C (2016) Unsupervised representation learning with deep convolutional generative adversarial networks. In International Conference on Learning Representations (ICLR)
[7]
Warde-Farley D, Yoshua B (2017) Improving generative adversarial networks with denoising feature matching. In International Conference on Learning Representations (ICLR)
[8]
Tim S, Ian G, Wojciech Z, Vicki C, Alec R, and Xi C Improved techniques for training gans Advances in neural information processing systems 2016 Berlin Springer 2234-2242
[9]
Tu DN, Trung L, Hung V, and Dinh P Dual discriminator generative adversarial nets Advances in neural information processing systems (NIPS) 2017 Berlin Springer 29
[10]
Tolstikhin IO et al. Adagan: boosting generative models Advances in neural information processing systems 2017 Berlin Springer 30
[11]
Wang C et al. Evolutionary generative adversarial networks IEEE Trans Evolut Comput 2019 23 6 921-934
[12]
Heusel M et al. Gans trained by a two time-scale update rule converge to a local nash equilibrium Advances in neural information processing systems 2017 Berlin Springer 30
[13]
Salimans T et al. Improved techniques for training gans Advances in neural information processing systems 2016 Berlin Springer 29
[14]
Szegedy C et al (2015) Going deeper with convolutions. In: Proceedings of the IEEE conference on computer vision and pattern recognition
[15]
Arjovsky M, Chintala S, Bottou L (2017) Wasserstein generative adversarial networks. In: International conference on machine learning, PMLR
Index Terms
- Improving generative adversarial network with multiple generators by evolutionary algorithms
Index terms have been assigned to the content through auto-classification.
Recommendations
Attentive evolutionary generative adversarial network
AbstractGenerative adversarial network (GAN) is an effective method to learn generative models from real data. But there are some drawbacks such as instability, mode collapse and low computational efficiency in the existing GANs. In this paper, attentive ...
Using holey fitness landscapes to counteract premature convergence in evolutionary algorithms
GECCO '08: Proceedings of the 10th annual conference companion on Genetic and evolutionary computationPremature convergence is a persisting problem in evolutionary optimisation, in particular - genetic algorithms. While a number of methods exist to approach this issue, they usually require problem specific calibration or only partially resolve the issue,...
An Improved Algorithm for Solving Scheduling Problems by Combining Generative Adversarial Network with Evolutionary Algorithms
CSAE '19: Proceedings of the 3rd International Conference on Computer Science and Application EngineeringWith1 the continuous application of evolutionary algorithms in various combinatorial optimization problems, the traditional evolutionary algorithms are prone to premature convergence and fall into local optimization solutions as the complexity of the ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
© International Society of Artificial Life and Robotics (ISAROB) 2022. Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
Publisher
Springer-Verlag
Berlin, Heidelberg
Publication History
Published: 01 November 2022
Accepted: 01 September 2022
Received: 10 June 2022
Author Tags
Qualifiers
- Research-article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 20 Dec 2024