ABSTRACT
Before the advent of Generative Adversarial Networks (GANs), Evolutionary Computation approaches made up the majority of the state of the art for image generation. Adversarial models employing Deep Convolutional Neural Networks better fitted for GPU computing have been at this point more efficient. Nevertheless, motivated by recent successes in GPU-accelerated Genetic Programming (GP) and given the disposition of expression-based solutions towards image evolution, we believe in the prospect of using symbolic expressions as generators for GANs, instead of neural networks. In this paper, we propose a novel GAN model called TGPGAN, where the traditional convolutional generator network is replaced with a GP approach. The generator iteratively evolves a population of expressions that are then passed to the discriminator module along with real images for backpropagation. Our experimental results show that it is possible to achieve comparable results to a typical Deep Convolutional GAN while benefiting from the flexibility enabled by an expression-based genotype. Moreover, this work serves as a proof of concept for the evolution of symbolic expressions within adversarial models.
- Martin Arjovsky, Soumith Chintala, and Léon Bottou. 2017. Wasserstein generative adversarial networks. In International conference on machine learning. PMLR, 214--223.Google Scholar
- Francisco Baeta, João Correia, Tiago Martins, and Penousal Machado. 2021. Speed benchmarking of genetic programming frameworks. In GECCO 2021 - Proceedings of the 2021 Genetic and Evolutionary Computation Conference. ACM, to appear.Google ScholarDigital Library
- Francisco Baeta, João Correia, Tiago Martins, and Penousal Machado. 2021. TensorGP --- Genetic Programming Engine in TensorFlow. In Applications of Evolutionary Computation - 24th International Conference, EvoApplications 2021. Springer, 763--778.Google ScholarDigital Library
- Lawrence Cayton. 2005. Algorithms for manifold learning. Univ. of California at San Diego Tech. Rep 12, 1--17 (2005), 1.Google Scholar
- João Correia, Penousal Machado, Juan Romero, Pedro Martins, and F Amílcar Cardoso. 2019. Breaking the Mould An Evolutionary Quest for Innovation Through Style Change. In Computational Creativity. Springer, 353--398.Google Scholar
- Victor Costa, Nuno Lourenço, João Correia, and Penousal Machado. 2019. COEGAN: Evaluating the coevolution effect in generative adversarial networks. In Proceedings of the genetic and evolutionary computation conference. 374--382.Google ScholarDigital Library
- Victor Costa, Nuno Lourenço, João Correia, and Penousal Machado. 2021. Demonstrating the Evolution of GANs through t-SNE. arXiv preprint arXiv:2102.00524 (2021).Google Scholar
- Erlend Gjesteland Ekern and Björn Gambäck. 2021. Interactive, Efficient and Creative Image Generation Using Compositional Pattern-Producing Networks. In International Conference on Computational Intelligence in Music, Sound, Art and Design (Part of EvoStar). Springer, 131--146.Google Scholar
- Philip Galanter. 2003. What is generative art? Complexity theory as a context for art theory. In In GA2003--6th Generative Art Conference. Citeseer.Google Scholar
- Ian Goodfellow. 2016. Nips 2016 tutorial: Generative adversarial networks. arXiv preprint arXiv:1701.00160 (2016).Google Scholar
- Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. Generative adversarial nets. Advances in neural information processing systems 27 (2014).Google Scholar
- David Ha. 2016. Generating Abstract Patterns with Tensor-Flow. blog.otoro.net (2016). https://blog.otoro.net/2016/03/25/generating-abstract-patterns-with-tensorflow/Google Scholar
- Penousal Machado and Amílcar Cardoso. 2002. All the truth about NEvAr. Applied Intelligence 16, 2 (2002), 101--118.Google ScholarDigital Library
- Tiago Martins, João Correia, Ernesto Costa, and Penousal Machado. 2019. Evolving Stencils for Typefaces: Combining Machine Learning, User's Preferences and Novelty. Complexity 2019 (2019).Google Scholar
- Luke Metz and Ishaan Gulrajani. 2017. Compositional pattern producing GAN. In NeurIPS Workshops, Vol. 1.Google Scholar
- Karl Sims. 1991. Artificial evolution for computer graphics. In Proceedings of the 18th annual conference on Computer graphics and interactive techniques. 319--328.Google ScholarDigital Library
- Kenneth O Stanley. 2007. Compositional pattern producing networks: A novel abstraction of development. Genetic programming and evolvahle machines 8, 2 (2007), 131--162.Google Scholar
- Kenneth O Stanley and Risto Miikkulainen. 2002. Evolving neural networks through augmenting topologies. Evolutionary computation 10, 2 (2002), 99--127.Google Scholar
- Chaoyue Wang, Chang Xu, Xin Yao, and Dacheng Tao. 2019. Evolutionary generative adversarial networks. IEEE Transactions on Evolutionary Computation 23, 6 (2019), 921--934.Google ScholarDigital Library
- Jun-Yan Zhu, Taesung Park, Phillip Isola, and Alexei A Efros. 2017. Unpaired image-to-image translation using cycle-consistent adversarial networks. In Proceedings of the IEEE international conference on computer vision. 2223--2232.Google ScholarCross Ref
Index Terms
- TGPGAN: towards expression-based generative adversarial networks
Recommendations
Exploring expression-based generative adversarial networks
GECCO '22: Proceedings of the Genetic and Evolutionary Computation Conference CompanionBefore the advent of Generative Adversarial Networks (GANs) in 2014, Evolutionary Computation approaches made up the majority of the state of art for image generation. Such approaches have been replaced with adversarial models using deep convolutional ...
CapsuleGAN: Generative Adversarial Capsule Network
Computer Vision – ECCV 2018 WorkshopsAbstractWe present Generative Adversarial Capsule Network (CapsuleGAN), a framework that uses capsule networks (CapsNets) instead of the standard convolutional neural networks (CNNs) as discriminators within the generative adversarial network (GAN) ...
Detect and Remove Watermark in Deep Neural Networks via Generative Adversarial Networks
Information SecurityAbstractDeep neural networks (DNN) have achieved remarkable performance in various fields. However, training a DNN model from scratch requires expensive computing resources and a lot of training data, which are difficult to obtain for most individual ...
Comments