Corpus ID: 211259510

A Group-Theoretic Framework for Data Augmentation

@article{Chen2019AGF,
  title={A Group-Theoretic Framework for Data Augmentation},
  author={Shuxiao Chen and Edgar Dobriban and J. Lee},
  journal={arXiv: Machine Learning},
  year={2019}
}
  • Shuxiao Chen, Edgar Dobriban, J. Lee
  • Published 2019
  • Mathematics, Computer Science
  • arXiv: Machine Learning
  • Data augmentation is a widely used trick when training deep neural networks: in addition to the original data, properly transformed data are also added to the training set. However, to the best of our knowledge, a clear mathematical framework to explain the performance benefits of data augmentation is not available. In this paper, we develop such a theoretical framework. We show data augmentation is equivalent to an averaging operation over the orbits of a certain group that keeps the data… CONTINUE READING
    7 Citations
    On the Benefits of Invariance in Neural Networks
    • 5
    • Highly Influenced
    • PDF
    On the Generalization Effects of Linear Transformations in Data Augmentation
    • 5
    • PDF
    UniformAugment: A Search-free Probabilistic Data Augmentation Approach
    • 2
    • Highly Influenced
    • PDF
    Implicit Regularization of Normalization Methods
    • 8
    Enhanced Convolutional Neural Tangent Kernels
    • 29
    • PDF