Examining Effects of Class Imbalance on Conditional GAN Training
Examining Effects of Class Imbalance on Conditional GAN Training
复制标题
检查类不平衡对条件 GAN 训练的影响
DOI：
10.1007/978-3-031-42505-9_40
复制
发表时间：
2023
期刊：
影响因子：
--
通讯作者：
Angryk R. A.
中科院分区：
文献类型：
--
作者：
Chen Y.;Kempton D. J.;Angryk R. A.
在这项工作中，我们研究了类不平衡对条件生成对抗网络(CGAN)模型生成的合成样本的准确性和多样性的影响。虽然许多利用GANS的研究在产生逼真的图像样本方面取得了非凡的成功，但这些研究通常假设使用经过良好处理和平衡的基准图像数据集，包括MNIST和CIFAR-10。然而，在现实世界的应用中，平衡良好的数据并不常见，例如检测欺诈、诊断糖尿病和预测太阳耀斑。众所周知，当类别标签分布不均匀时，分类算法的预测能力会受到严重影响，这一现象被称为“类别不平衡问题”。我们表明，训练集的不平衡也会影响CGAN模型的样本生成。我们利用著名的MNIST数据集，通过采样来控制数据中某些类别的不平衡比例。我们能够证明，在存在类别失衡的情况下，生成的样本的质量和多样性都会受到影响，并提出了一种名为两阶段CGAN的新框架，以在这种情况下产生高质量的合成样本。我们的结果表明，与用于类失衡修复的典型过采样和欠采样技术相比，所提出的框架具有显著的改进。
In this work, we investigate the impact of class imbalance on the accuracy and diversity of synthetic samples generated by conditional generative adversarial networks (CGAN) models. Though many studies utilizing GANs have seen extraordinary success in producing realistic image samples, these studies generally assume the use of well-processed and balanced benchmark image datasets, including MNIST and CIFAR-10. However, well-balanced data is uncommon in real world applications such as detecting fraud, diagnosing diabetes, and predicting solar flares. It is well known that when class labels are not distributed uniformly, the predictive ability of classification algorithms suffers significantly, a phenomenon known as the “class-imbalance problem.” We show that the imbalance in the training set can also impact sample generation of CGAN models. We utilize the well known MNIST datasets, controlling the imbalance ratio of certain classes within the data through sampling. We are able to show that both the quality and diversity of generated samples suffer in the presence of class imbalances and propose a novel framework named Two-stage CGAN to produce high-quality synthetic samples in such cases. Our results indicate that the proposed framework provides a significant improvement over typical oversampling and undersampling techniques utilized for class imbalance remediation.