Corpus ID: 210921097

Imperfect ImaGANation: Implications of GANs Exacerbating Biases on Facial Data Augmentation and Snapchat Selfie Lenses

@article{Jain2020ImperfectII,
  title={Imperfect ImaGANation: Implications of GANs Exacerbating Biases on Facial Data Augmentation and Snapchat Selfie Lenses},
  author={Niharika Jain and Alberto Olmo Hernandez and Sailik Sengupta and L. Manikonda and S. Kambhampati},
  journal={ArXiv},
  year={2020},
  volume={abs/2001.09528}
}
  • Niharika Jain, Alberto Olmo Hernandez, +2 authors S. Kambhampati
  • Published 2020
  • Computer Science, Engineering, Mathematics
  • ArXiv
  • Recently, the use of synthetic data generated by GANs has become a popular method to do data augmentation for many applications. While practitioners celebrate this as an economical way to obtain synthetic data for training data-hungry machine learning models, it is not clear that they recognize the perils of such an augmentation technique when applied to an already-biased dataset. Although one expects GANs to replicate the distribution of the original data, in real-world settings with limited… CONTINUE READING
    7 Citations

    Figures from this paper

    Improving the Fairness of Deep Generative Models without Retraining
    • 1
    • PDF
    Exploring Racial Bias within Face Recognition via per-subject Adversarially-Enabled Data Augmentation
    • 2
    • PDF
    Directional Bias Amplification
    • PDF
    DAPT 2020-Constructing a Benchmark Dataset for Advanced Persistent Threats
    • PDF
    Transdisciplinary AI Observatory - Retrospective Analyses and Future-Oriented Contradistinctions
    • PDF

    References

    SHOWING 1-10 OF 67 REFERENCES
    Data Augmentation Generative Adversarial Networks
    • 394
    • PDF
    A Style-Based Generator Architecture for Generative Adversarial Networks
    • 1,583
    • PDF
    Mode Regularized Generative Adversarial Networks
    • 313
    • PDF
    Do GANs actually learn the distribution? An empirical study
    • 133
    • PDF
    Generalization of Deep Neural Networks for Chest Pathology Classification in X-Rays Using Generative Adversarial Networks
    • 95
    • PDF