pith. sign in

arxiv: 2606.26169 · v1 · pith:DIJETEYDnew · submitted 2026-06-24 · 💻 cs.LG · cs.AI

Neural Architecture Search for Generative Adversarial Networks: A Comprehensive Review and Critical Analysis

Abrar Alotaibi , Moataz Ahmed This is my paper

Pith reviewed 2026-06-26 01:57 UTC · model grok-4.3

classification 💻 cs.LG cs.AI
keywords neural architecture searchgenerative adversarial networksevolutionary algorithmsgradient-based methodsevaluation metricsInception ScoreFréchet Inception DistanceGAN stability
0
0 comments X

The pith

NAS for GANs works best with evolutionary algorithms and gradient-based search when evaluation uses metrics beyond IS and FID plus diverse datasets.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This review organizes existing work on using neural architecture search to design Generative Adversarial Networks automatically. It groups methods by search strategy, the metrics used to judge results, and the performance numbers reported, then compares outcomes across those groups. The central finding is that evolutionary algorithms and gradient-based techniques show advantages for stability and efficiency in the cases examined. The paper stresses that standard scores like Inception Score and Fréchet Inception Distance miss important weaknesses and that tests on narrow datasets give incomplete pictures. The structured comparison is meant to help researchers pick approaches and identify where new methods are still needed.

Core claim

The paper establishes that NAS improves GAN performance, stability, and efficiency over manual design, with evolutionary algorithms and gradient-based methods proving superior in certain contexts. It shows that robust evaluation requires metrics beyond IS and FID and that performance claims need testing on diverse datasets rather than standard benchmarks alone. The categorization by search strategies, metrics, and outcomes supplies a map for comparing techniques and spotting gaps.

What carries the argument

Categorization of NAS-GAN approaches according to search strategies, evaluation metrics, and performance outcomes

If this is right

  • Researchers should favor evolutionary or gradient-based NAS when building new GAN architectures for tasks where stability matters.
  • Evaluation protocols for NAS-GANs must incorporate metrics that capture failure modes missed by IS and FID.
  • Benchmarking GAN performance requires datasets that vary in domain, size, and complexity rather than relying on a few standard collections.
  • New NAS methods for GANs should target the stability and efficiency gains already observed in the reviewed evolutionary and gradient approaches.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same search-strategy comparison could be applied to NAS for diffusion models or other generative families to test whether evolutionary methods generalize.
  • Adoption of broader metrics might change which architectures rank highest, exposing cases where current top performers fail on real deployment criteria.
  • Diverse-dataset testing could reveal that some NAS-found GANs overfit to narrow benchmarks and underperform when data distributions shift.

Load-bearing premise

The papers chosen for the review represent the full range of NAS-GAN work and the chosen grouping criteria reveal the real differences between methods.

What would settle it

A follow-up survey that adds many more NAS-GAN papers and finds different search strategies superior or demonstrates that IS and FID alone reliably predict real-world GAN behavior would undermine the review's conclusions.

Figures

Figures reproduced from arXiv: 2606.26169 by Abrar Alotaibi, Moataz Ahmed.

Figure 1
Figure 1. Figure 1: Literature distribution. We employed the following search strings to identify relevant studies: • In Google Scholar, IEEE, ACM, and Springer: (“Generative Adversarial Network” OR “GAN*”) AND (“Architecture” OR “Architectural”) AND (“Search” OR “Op￾timization”) AND (“Reinforcement Learning” OR “Policy” OR “Evolutionary” OR “Evolutionary Algorithm*” OR “Genetic Algorithm*” OR “Differential” OR “Gradient￾base… view at source ↗
Figure 2
Figure 2. Figure 2: Distribution of NAS-GAN Methods by Search Strategy. 4.1.1. Evolutionary Algorithms Approaches EA is a family of metaheuristic optimization algorithms, inspired by collective behav￾iors observed in nature [21]. These algorithms can navigate complex high-dimensional search spaces to find near-optimal solutions, making them well-suited for automat￾ically finding answers to the many interdependent architectura… view at source ↗
Figure 3
Figure 3. Figure 3: Distribution of Search Space Types in NAS-GAN Literature. 4.2.1. Entire/Chain-Structure Space Entire/chain-structure neural networks represent one of the simplest search spaces. In this space, an architecture can be described as a sequence of multiple layers, where the output of a layer serves as input for the next layer. NSGA-II DCGAN [28] presents an entire/chain -structure search space. This search spac… view at source ↗
read the original abstract

Neural Architecture Search (NAS) has emerged as a pivotal technique in optimizing the design of Generative Adversarial Networks (GANs), automating the search for effective architectures while addressing the challenges inherent in manual design. This paper provides a comprehensive review of NAS methods applied to GANs, categorizing and comparing various approaches based on criteria such as search strategies, evaluation metrics, and performance outcomes. The review highlights the benefits of NAS in improving GAN performance, stability, and efficiency, while also identifying limitations and areas for future research. Key findings include the superiority of evolutionary algorithms and gradient-based methods in certain contexts, the importance of robust evaluation metrics beyond traditional scores like Inception Score (IS) and Fr\'echet Inception Distance (FID), and the need for diverse datasets in assessing GAN performance. By presenting a structured comparison of existing NAS-GAN techniques, this paper aims to guide researchers in developing more effective NAS methods and advancing the field of GANs.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The paper provides a comprehensive review of Neural Architecture Search (NAS) methods applied to Generative Adversarial Networks (GANs). It categorizes and compares approaches based on search strategies, evaluation metrics, and performance outcomes; highlights benefits of NAS for GAN performance, stability, and efficiency; identifies limitations; and outlines future research. Key findings asserted include the superiority of evolutionary algorithms and gradient-based methods in certain contexts, the importance of robust evaluation metrics beyond IS and FID, and the need for diverse datasets in assessing GAN performance.

Significance. If the surveyed papers form a representative sample and the chosen categorization axes meaningfully distinguish approaches, the review could usefully guide researchers by synthesizing the NAS-GAN literature and emphasizing evaluation challenges. The absence of any stated search protocol, inclusion criteria, database, date range, or paper count, however, prevents assessment of whether the superiority and metric recommendations are grounded in the broader literature rather than selection artifacts.

major comments (2)
  1. [Abstract] Abstract: the abstract states key findings on superiority of evolutionary/gradient-based methods and metric recommendations but provides no details on paper selection criteria, coverage of the literature, total papers reviewed, search protocol, databases, keywords, or date range; without this information the central claims cannot be verified as representative.
  2. [Introduction] Introduction (and any methods or survey-design section): the review assumes the selected papers are representative and that the categorization criteria (search strategies, metrics, outcomes) capture meaningful differences, yet no justification, inclusion/exclusion criteria, or validation of the taxonomy is supplied; this directly undermines the reported patterns of superiority 'in certain contexts'.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments regarding the transparency of our survey methodology. We address each major comment below and will perform a major revision to incorporate the requested details.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the abstract states key findings on superiority of evolutionary/gradient-based methods and metric recommendations but provides no details on paper selection criteria, coverage of the literature, total papers reviewed, search protocol, databases, keywords, or date range; without this information the central claims cannot be verified as representative.

    Authors: We agree that the abstract lacks these details and that they are needed to support the central claims. In the revised manuscript we will expand the abstract to note the approximate number of papers reviewed and the overall search scope, while adding a new 'Survey Methodology' subsection (placed after the Introduction) that fully specifies the databases, keywords, date range, and inclusion criteria. revision: yes

  2. Referee: [Introduction] Introduction (and any methods or survey-design section): the review assumes the selected papers are representative and that the categorization criteria (search strategies, metrics, outcomes) capture meaningful differences, yet no justification, inclusion/exclusion criteria, or validation of the taxonomy is supplied; this directly undermines the reported patterns of superiority 'in certain contexts'.

    Authors: We accept that the manuscript currently provides no explicit justification or criteria. We will add a dedicated subsection that justifies the taxonomy (search strategy, metrics, outcomes) by reference to prior NAS and GAN surveys, states the inclusion/exclusion criteria, and notes the total papers examined. This will ground the observed patterns in the selected corpus without claiming exhaustive coverage of the entire literature. revision: yes

Circularity Check

0 steps flagged

Review paper with no internal derivation chain or self-referential reductions

full rationale

This is a survey paper whose central content consists of categorizing and summarizing external literature on NAS-GAN methods. The abstract and described structure contain no equations, predictions, fitted parameters, or derivations that could reduce to quantities defined within the paper itself. Key findings are explicitly attributed to surveyed prior works rather than generated internally. No self-citation load-bearing steps, ansatzes, or uniqueness claims appear. The paper is therefore self-contained against external benchmarks with no circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The paper is a literature review and introduces no original mathematical models, parameters, or entities; all content rests on the body of prior NAS-GAN publications it cites.

pith-pipeline@v0.9.1-grok · 5693 in / 1116 out tokens · 18105 ms · 2026-06-26T01:57:12.422763+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

60 extracted references · 7 canonical work pages

  1. [1]

    A survey on Image Data Augmentation for Deep Learning.J

    Shorten, C.; Khoshgoftaar, T.M. A survey on Image Data Augmentation for Deep Learning.J. Big Data2019,6, 60

  2. [2]

    Fine-tuning Generative Adversarial Networks using Metaheuristics

    Souza, L.A.; Passos, L.A.; Mendel, R.; Ebigbo, A.; Probst, A.; Messmann, H.; Palm, C.; Papa, J.P . Fine-tuning Generative Adversarial Networks using Metaheuristics. In Proceedings of the Bildverarbeitung für die Medizin 2021, Regensburg, Germany, 7–9 March 2021; Springer: Wiesbaden, Germany, 2021; pp. 205–210

    2021

  3. [3]

    Advanced metaheuristic optimization techniques in applications of deep neural networks: A review.Neural Comput

    Abd Elaziz, M.; Dahou, A.; Abualigah, L.; Yu, L.; Alshinwan, M.; Khasawneh, A.M.; Lu, S. Advanced metaheuristic optimization techniques in applications of deep neural networks: A review.Neural Comput. Appl.2021,33, 14079–14099

    2021

  4. [4]

    Applications of Generative Adversarial Networks (GANs) in Positron Emission Tomography (PET) imaging: A review.Eur

    Apostolopoulos, I.; Papathanasiou, N.; Apostolopoulos, D.; Panayiotakis, G. Applications of Generative Adversarial Networks (GANs) in Positron Emission Tomography (PET) imaging: A review.Eur. J. Nucl. Med. Mol. Imaging2022,49, 1–23

  5. [5]

    GAN-based anomaly detection: A review.Neurocomputing2022, 493, 497–535

    Xia, X.; Pan, X.; Li, N.; He, X.; Ma, L.; Zhang, X.; Ding, N. GAN-based anomaly detection: A review.Neurocomputing2022, 493, 497–535

  6. [6]

    StyleMC: Multi-channel based fast text-guided image generation and manipula- tion

    Kocasari, U.; Dirik, A.; Tiftikci, M.; Yanardag, P . StyleMC: Multi-channel based fast text-guided image generation and manipula- tion. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, Waikoloa, HI, USA, 3–7 January 2022; pp. 895–904. Appl. Sci.2025,1, 0 30 of 32

    2022

  7. [7]

    Text augmentation for neural networks

    Mosolova, A.V .; Fomin, V .V .; Bondarenko, I.Y. Text augmentation for neural networks. In Proceedings of the CEUR Workshop Proceedings, 2018; Volume 2268, pp. 104–109

    2018

  8. [8]

    Optimization of deep neural networks: A survey and unified taxonomy.arXiv2020, arXiv:2006.05597

    Talbi, E.G. Optimization of deep neural networks: A survey and unified taxonomy.arXiv2020, arXiv:2006.05597

    arXiv 2006

  9. [9]

    Catastrophic forgetting and mode collapse in GANs

    Thanh-Tung, H.; Tran, T. Catastrophic forgetting and mode collapse in GANs. In Proceedings of the 2020 International Joint Conference on Neural Networks (IJCNN), Glasgow, UK, 19–24 July 2020; pp. 1–10

    2020

  10. [10]

    Neural architecture search: A survey.J

    Elsken, T.; Metzen, J.H.; Hutter, F. Neural architecture search: A survey.J. Mach. Learn. Res.2019,20, 1–21

    2019

  11. [11]

    AutoML: A survey of the state-of-the-art.Knowl.-Based Syst.2021,212, 106622

    He, X.; Zhao, K.; Chu, X. AutoML: A survey of the state-of-the-art.Knowl.-Based Syst.2021,212, 106622

    2021

  12. [12]

    AutoGAN: Neural architecture search for generative adversarial networks

    Gong, X.; Chang, S.; Jiang, Y.; Wang, Z. AutoGAN: Neural architecture search for generative adversarial networks. In Proceedings of the IEEE/CVF International Conference on Computer Vision, Seoul, Republic of Korea, 27 October–2 November 2019; pp. 3224–3234

    2019

  13. [13]

    COEGAN: Evaluating the coevolution effect in generative adversarial networks

    Costa, V .; Lourenço, N.; Correia, J.; Machado, P . COEGAN: Evaluating the coevolution effect in generative adversarial networks. In Proceedings of the Genetic and Evolutionary Computation Conference, Prague, Czech Republic, 13–17 July 2019; pp. 374–382

    2019

  14. [14]

    AlphaGAN: Fully differentiable architecture search for generative adversarial networks

    Tian, Y.; Shen, L.; Su, G.; Li, Z.; Liu, W. AlphaGAN: Fully differentiable architecture search for generative adversarial networks. IEEE Trans. Pattern Anal. Mach. Intell.2021,44, 6752–6766

    2021

  15. [15]

    Automating generative adversarial networks using neural architecture search: A review

    Ganepola, V .V .V .; Wirasingha, T. Automating generative adversarial networks using neural architecture search: A review. In Proceedings of the 2021 International Conference on Emerging Smart Computing and Informatics (ESCI), Pune, India, 5–7 March 2021; pp. 577–582

    2021

  16. [16]

    Neural Architecture Search for Generative Adversarial Networks: A Review

    Buthgamumudalige, V .U.; Wirasingha, T. Neural Architecture Search for Generative Adversarial Networks: A Review. In Proceedings of the 2021 10th International Conference on Information and Automation for Sustainability (ICIAfS), Negambo, Sri Lanka, 11–13 August 2021; pp. 246–251

    2021

  17. [17]

    The application of evolutionary computation in generative adversarial networks (GANs): A systematic literature survey.Artif

    Wang, Y.; Zhang, Q.; Wang, G.G.; Cheng, H. The application of evolutionary computation in generative adversarial networks (GANs): A systematic literature survey.Artif. Intell. Rev.2024,57, 182. https://doi.org/10.1007/s10462-024-10818-y

    work page doi:10.1007/s10462-024-10818-y 2024

  18. [18]

    Neural Architecture Search Survey: A Computer Vision Perspective.Sensors2023,23, 1713

    Kang, J.-S.; Kang, J.; Kim, J.-J.; Jeon, K.-W.; Chung, H.-J.; Park, B.-H. Neural Architecture Search Survey: A Computer Vision Perspective.Sensors2023,23, 1713

  19. [19]

    Neural architecture search: Insights from 1000 papers.arXiv2023, arXiv:2301.08727

    White, C.; Safari, M.; Sukthanker, R.; Ru, B.; Elsken, T.; Zela, A.; Dey, D.; Hutter, F. Neural architecture search: Insights from 1000 papers.arXiv2023, arXiv:2301.08727

    arXiv

  20. [20]

    Kitchenham, B.Procedures for Performing Systematic Reviews; Technical report; Keele University: Newcastle, UK, 2004

    2004

  21. [21]

    Talbi, E.G.Metaheuristics: From Design to Implementation; John Wiley & Sons: Hoboken, NJ, USA, 2009; Volume 74

    2009

  22. [22]

    Evolutionary generative adversarial networks.IEEE Trans

    Wang, C.; Xu, C.; Yao, X.; Tao, D. Evolutionary generative adversarial networks.IEEE Trans. Evol. Comput.2019,23, 921–934

    2019

  23. [23]

    Towards distributed coevolutionary GANs.arXiv2018, arXiv:1807.08194

    Al-Dujaili, A.; Schmiedlechner, T.; Hemberg, E.; O’Reilly, U.M. Towards distributed coevolutionary GANs.arXiv2018, arXiv:1807.08194

    Pith/arXiv arXiv

  24. [24]

    Spatial Evolutionary Generative Adversarial Networks

    Toutouh, J.; Hemberg, E.; O’Reilly, U.M. Spatial Evolutionary Generative Adversarial Networks. In Proceedings of the Genetic and Evolutionary Computation Conference, Prague, Czech Republic, 13–17 July 2019; pp. 472–480

    2019

  25. [25]

    Evolved GANs for generating pareto set approximations

    Garciarena, U.; Santana, R.; Mendiburu, A. Evolved GANs for generating pareto set approximations. In Proceedings of the Genetic and Evolutionary Computation Conference, GECCO ’18, Kyoto, Japan, 15–19 July 2018; pp. 434–441. https: //doi.org/10.1145/3205455.3205550

    work page doi:10.1145/3205455.3205550 2018

  26. [26]

    Autonomously and simultaneously refining deep neural network parameters by a bi-generative adversarial network aided genetic algorithm.arXiv2018, arXiv:1809.10244

    Lu, Y.; Kakillioglu, B.; Velipasalar, S. Autonomously and simultaneously refining deep neural network parameters by a bi-generative adversarial network aided genetic algorithm.arXiv2018, arXiv:1809.10244

    Pith/arXiv arXiv

  27. [27]

    3D ShapeNets: A deep representation for volumetric shapes

    Wu, Z.; Song, S.; Khosla, A.; Yu, F.; Zhang, L.; Tang, X.; Xiao, J. 3D ShapeNets: A deep representation for volumetric shapes. In Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA, 7–12 June 2015; pp. 1912–1920. https://doi.org/10.1109/CVPR.2015.7298801

    work page doi:10.1109/cvpr.2015.7298801 2015

  28. [28]

    Structure tuning method on deep convolutional generative adversarial network with nondomi- nated sorting genetic algorithm II.Concurr

    Du, L.; Cui, Z.; Wang, L.; Ma, J. Structure tuning method on deep convolutional generative adversarial network with nondomi- nated sorting genetic algorithm II.Concurr. Comput. Pract. Exp.2020,32, e5688

    2020

  29. [29]

    Evolutionary Architectural Search for Generative Adversarial Networks.IEEE Trans

    Lin, Q.; Fang, Z.; Chen, Y.; Tan, K.C.; Li, Y. Evolutionary Architectural Search for Generative Adversarial Networks.IEEE Trans. Emerg. Top. Comput. Intell.2022,6, 783–794

    2022

  30. [30]

    EAGAN: Efficient two-stage evolutionary architecture search for GANs

    Ying, G.; He, X.; Gao, B.; Han, B.; Chu, X. EAGAN: Efficient two-stage evolutionary architecture search for GANs. In Proceedings of the European Conference on Computer Vision, Tel Aviv, Israel, 23–27 October 2022; Springer: Cham, Switzerland, 2022; pp. 37–53

    2022

  31. [31]

    Evolutionary Architecture Search for Generative Adversarial Networks Based on Weight Sharing.IEEE Trans

    Xue, Y.; Tong, W.; Neri, F.; Chen, P .; Luo, T.; Zhen, L.; Wang, X. Evolutionary Architecture Search for Generative Adversarial Networks Based on Weight Sharing.IEEE Trans. Evol. Comput.2024,28, 653–667

    2024

  32. [32]

    A Multi-Objective Architecture Search for Generative Adversarial Networks

    Kobayashi, M.; Nagao, T. A Multi-Objective Architecture Search for Generative Adversarial Networks. In Proceedings of the 2020 Genetic and Evolutionary Computation Conference Companion, Cancún, Mexico, 8–12 July 2020; pp. 133–134

    2020

  33. [33]

    Sutton, R.S.; Barto, A.G.Reinforcement Learning: An Introduction; MIT Press: Cambridge, MA, USA, 2018

    2018

  34. [34]

    AGAN: Towards automated design of generative adversarial networks.arXiv2019, arXiv:1906.11080

    Wang, H.; Huan, J. AGAN: Towards automated design of generative adversarial networks.arXiv2019, arXiv:1906.11080. Appl. Sci.2025,1, 0 31 of 32

    Pith/arXiv arXiv 1906

  35. [35]

    Searching Towards Class-Aware Generators for Conditional Generative Adversarial Networks

    Zhou, P .; Xie, L.; Ni, B.; Tian, Q. Searching Towards Class-Aware Generators for Conditional Generative Adversarial Networks. IEEE Signal Process. Lett.2022,29, 1669–1673. https://doi.org/10.1109/LSP .2022.3193589

    work page doi:10.1109/lsp 2022

  36. [36]

    Off-policy reinforcement learning for efficient and effective gan architecture search

    Tian, Y.; Wang, Q.; Huang, Z.; Li, W.; Dai, D.; Yang, M.; Wang, J.; Fink, O. Off-policy reinforcement learning for efficient and effective gan architecture search. In Proceedings of the Computer Vision—ECCV 2020: 16th European Conference, Glasgow, UK, 23–28 August 2020; Proceedings, Part VII 16; Springer: Berlin/Heidelberg, Germany, 2020; pp. 175–192

    2020

  37. [37]

    Neural Architecture Search With a Lightweight Transformer for Text-to-Image Synthesis.IEEE Trans

    Li, W.; Wen, S.; Shi, K.; Yang, Y.; Huang, T. Neural Architecture Search With a Lightweight Transformer for Text-to-Image Synthesis.IEEE Trans. Netw. Sci. Eng.2022,9, 1567–1576. https://doi.org/10.1109/TNSE.2022.3147787

    work page doi:10.1109/tnse.2022.3147787 2022

  38. [38]

    Goodfellow, I.; Bengio, Y.; Courville, A.Deep Learning; MIT Press: Cambridge, MA, USA, 2016

    2016

  39. [39]

    AdversarialNAS: Adversarial Neural Architecture Search for GANs

    Gao, C.; Chen, Y.; Liu, S.; Tan, Z.; Yan, S. AdversarialNAS: Adversarial Neural Architecture Search for GANs. In Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 14–19 June 2020; pp. 5679–5688

    2020

  40. [40]

    DEGAS: Differentiable efficient generator search.Neural Comput

    Doveh, S.; Giryes, R. DEGAS: Differentiable efficient generator search.Neural Comput. Appl.2021,33, 17173–17184

    2021

  41. [41]

    GAN Compression: Efficient Architectures for Interactive Conditional GANs

    Li, M.; Lin, J.; Ding, Y.; Liu, Z.; Zhu, J.Y.; Han, S. GAN Compression: Efficient Architectures for Interactive Conditional GANs. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 14–19 June 2020

    2020

  42. [42]

    Differentiable Architecture Search With Attention Mechanisms for Generative Adversarial Networks

    Xue, Y.; Chen, K.; Neri, F. Differentiable Architecture Search With Attention Mechanisms for Generative Adversarial Networks. IEEE Trans. Emerg. Top. Comput. Intell.2024,8, 3141–3151. https://doi.org/10.1109/TETCI.2024.3369998

    work page doi:10.1109/tetci.2024.3369998 2024

  43. [43]

    Improved techniques for training gans

    Salimans, T.; Goodfellow, I.; Zaremba, W.; Cheung, V .; Radford, A.; Chen, X. Improved techniques for training gans. In Proceedings of the Advances in Neural Information Processing Systems, Barcelona, Spain, 5–10 December 2016; Volume 29

    2016

  44. [44]

    Rethinking the Inception Architecture for Computer Vision

    Szegedy, C.; Vanhoucke, V .; Ioffe, S.; Shlens, J.; Wojna, Z. Rethinking the Inception Architecture for Computer Vision. In Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV , USA, 26 June–1 July 2016; pp. 2818–2826

    2016

  45. [45]

    Gans trained by a two time-scale update rule converge to a local nash equilibrium

    Heusel, M.; Ramsauer, H.; Unterthiner, T.; Nessler, B.; Hochreiter, S. Gans trained by a two time-scale update rule converge to a local nash equilibrium. In Proceedings of the Advances in Neural Information Processing Systems, Long Beach, CA, USA, 4–9 December 2017; Volume 30

    2017

  46. [46]

    Regularized evolution for image classifier architecture search

    Real, E.; Aggarwal, A.; Huang, Y.; Le, Q.V . Regularized evolution for image classifier architecture search. In Proceedings of the AAAI Conference on Artificial Intelligence, Honolulu, HI, USA, 27 January–1 February 2019; Volume 33, pp. 4780–4789

    2019

  47. [47]

    EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks

    Tan, M.; Le, Q. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. In Proceedings of the International Conference on Machine Learning, PMLR, Long Beach, CA, USA, 9–15 June 2019; pp. 6105–6114

    2019

  48. [48]

    Veegan: Reducing mode collapse in gans using implicit variational learning

    Srivastava, A.; Valkov, L.; Russell, C.; Gutmann, M.U.; Sutton, C. Veegan: Reducing mode collapse in gans using implicit variational learning. In Proceedings of the Advances in Neural Information Processing Systems, Long Beach, CA, USA, 4–9 December 2017; Volume 30

    2017

  49. [49]

    Which training methods for GANs do actually converge? In Proceedings of the International Conference on Machine Learning, PMLR, Stockholm, Sweden, 10–15 July 2018; pp

    Mescheder, L.; Geiger, A.; Nowozin, S. Which training methods for GANs do actually converge? In Proceedings of the International Conference on Machine Learning, PMLR, Stockholm, Sweden, 10–15 July 2018; pp. 3481–3490

    2018

  50. [50]

    The unreasonable effectiveness of deep features as a perceptual metric

    Zhang, R.; Isola, P .; Efros, A.A.; Shechtman, E.; Wang, O. The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 18–23 June 2018; pp. 586–595

    2018

  51. [51]

    The MNIST Database of Handwritten Digit Images for Machine Learning Research [Best of the Web].IEEE Signal Process

    Deng, L. The MNIST Database of Handwritten Digit Images for Machine Learning Research [Best of the Web].IEEE Signal Process. Mag.2012,29, 141–142

    2012

  52. [52]

    Deep learning face attributes in the wild

    Liu, Z.; Luo, P .; Wang, X.; Tang, X. Deep learning face attributes in the wild. In Proceedings of the IEEE International Conference on Computer Vision, Santiago, Chile, 7–13 December 2015; pp. 3730–3738

    2015

  53. [53]

    Lsun: Construction of a large-scale image dataset using deep learning with humans in the loop.arXiv2015, arXiv:1506.03365

    Yu, F.; Seff, A.; Zhang, Y.; Song, S.; Funkhouser, T.; Xiao, J. Lsun: Construction of a large-scale image dataset using deep learning with humans in the loop.arXiv2015, arXiv:1506.03365

    Pith/arXiv arXiv

  54. [54]

    Krizhevsky, A.Learning Multiple Layers of Features from Tiny Images; Technical report; University of Toronto: Toronto, ON, Canada, 2009

    2009

  55. [55]

    Generative adversarial nets

    Goodfellow, I.; Pouget-Abadie, J.; Mirza, M.; Xu, B.; Warde-Farley, D.; Ozair, S.; Courville, A.; Bengio, Y. Generative adversarial nets. In Proceedings of the Advances in Neural Information Processing Systems, Montreal, QC, Canada, 8–13 December 2014; Volume 27

    2014

  56. [56]

    Conditional generative adversarial nets.arXiv2014, arXiv:1411.1784

    Mirza, M.; Osindero, S. Conditional generative adversarial nets.arXiv2014, arXiv:1411.1784

    Pith/arXiv arXiv

  57. [57]

    Uncover This Tech Term: Generative Adversarial Networks.Korean J

    Ahmed, H.S. Uncover This Tech Term: Generative Adversarial Networks.Korean J. Radiol.2024,25, 493 – 498

    2024

  58. [58]

    Data augmentation using GANs.arXiv2019, arXiv:1904.09135

    Tanaka, F.H.K.D.S.; Aranha, C. Data augmentation using GANs.arXiv2019, arXiv:1904.09135

    Pith/arXiv arXiv 1904

  59. [59]

    IEEE Transactions on Pattern Analysis and Machine Intelli- gence43(12), 4217–4228 (2021).https://doi.org/10.1109/TPAMI.2020.2970919

    Karras, T.; Laine, S.; Aila, T. A Style-Based Generator Architecture for Generative Adversarial Networks.IEEE Trans. Pattern Anal. Mach. Intell.2021,43, 4217–4228. https://doi.org/10.1109/TPAMI.2020.2970919

    work page doi:10.1109/tpami.2020.2970919 2021

  60. [60]

    Progressive growing of gans for improved quality, stability, and variation.arXiv2018, arXiv:1710.10196

    Karras, T.; Aila, T.; Laine, S.; Lehtinen, J. Progressive growing of gans for improved quality, stability, and variation.arXiv2018, arXiv:1710.10196. Appl. Sci.2025,1, 0 32 of 32 Disclaimer/Publisher’s Note:The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or...

    Pith/arXiv arXiv 2025