Wasserstein Generative Adversarial Network with Gradient Penalty for Handwritten Digit Generation
Authors:
- 29 November 2024
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 375-379
Abstract
The Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) offers a powerful solution for generating high-quality synthetic data, addressing instability and mode collapse issues common in traditional GANs. This article explores the application of WGAN-GP in generating realistic handwritten digits using the MNIST dataset. WGAN-GP utilizes the Wasserstein distance metric rather than Jensen-Shannon divergence and introduces a gradient penalty to maintain the Lipschitz continuity constraint, which significantly improves training stability and image fidelity. We examine the effectiveness of WGAN-GP’s gradient penalty in preventing issues such as vanishing and exploding gradients, optimizing key hyperparameters, and enhancing the architecture for diverse and high-quality digit generation. Experimental results demonstrate WGAN-GP’s success in producing varied, lifelike handwritten digits and highlight the impact of key design choices, including latent space dimensionality and critic iterations, on image quality. Our findings suggest that WGAN-GP is a viable approach for applications requiring high-quality, synthetic digit datasets, potentially benefiting downstream supervised tasks in handwritten digit recognition and augmentation.Keywords
This publication has 17 references indexed in Scilit:
- Frequency-domain enhanced bi-directional recurrent quantum network for stock price trend predictionMultimedia Tools and Applications, 2023
- Evaluating Generative Adversarial Networks for particle hit generation in a cylindrical drift chamber using Fréchet Inception DistanceJournal of Instrumentation, 2023
- Multiscale attention dynamic aware network for fine‐grained visual categorizationElectronics Letters, 2022
- Dynamic manifold Boltzmann optimization based on self‐supervised learning for human motion estimationIET Image Processing, 2022
- Quality Assessment Method for GAN Based on Modified Metrics Inception Score and Fréchet Inception Distance2020
- Data augmentation in fault diagnosis based on the Wasserstein generative adversarial network with gradient penaltyNeurocomputing, 2020
- Image generation by GAN and style transfer for agar plate image segmentationComputer Methods and Programs in Biomedicine, 2020
- Conditional Wasserstein generative adversarial network-gradient penalty-based approach to alleviating imbalanced data classificationInformation Sciences, 2020
- Generative adversarial network in medical imaging: A reviewMedical Image Analysis, 2019
- An Analysis of Edge Detection by Using the Jensen-Shannon DivergenceJournal of Mathematical Imaging and Vision, 2000