mlsanityData to 7 Oct 2026

Observations on K-image Expansion of Image-Mixing Augmentation for Classification

arXiv 2110.04248 ↗First posted 8 Oct 2021cs.CV cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

Image-mixing augmentations (e.g., Mixup and CutMix), which typically involve mixing two images, have become the de-facto training techniques for image classification. Despite their huge success in image classification, the number of images to be mixed has not been elucidated in the literature: only the naive K-image expansion has been shown to lead to performance degradation. This study derives a new K-image mixing augmentation based on the stick-breaking process under Dirichlet prior distribution. We demonstrate the superiority of our K-image expansion augmentation over conventional two-image mixing augmentation methods through extensive experiments and analyses: (1) more robust and generalized classifiers; (2) a more desirable loss landscape shape; (3) better adversarial robustness. Moreover, we show that our probabilistic model can measure the sample-wise uncertainty and boost the efficiency for network architecture search by achieving a 7-fold reduction in the search time. Code will be available at https://github.com/yjyoo3312/DCutMix-PyTorch.git.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 45 of its 63 references that are in the map. First posted on arXiv 8 Oct 2021.

Builds on

45 of 63 references.

References · 45

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  3. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  4. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  5. 2013Intriguing properties of neural networkscited by 5,114
  6. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  7. 2016Wide Residual Networkscited by 3,057
  8. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
  9. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,540
  10. 2016Identity Mappings in Deep Residual Networkscited by 1,960
  11. 2018DARTS: Differentiable Architecture Searchcited by 1,782
  12. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,623
33 more
  1. 2016Neural Architecture Search with Reinforcement Learningcited by 1,489
  2. 2019Natural Adversarial Examplescited by 1,401
  3. 2016On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minimacited by 1,337
  4. 2017Learning Transferable Architectures for Scalable Image Recognitioncited by 1,097
  5. 2016Deep Networks with Stochastic Depthcited by 973
  6. 2018MnasNet: Platform-Aware Neural Architecture Search for Mobilecited by 937
  7. 2018Regularized Evolution for Image Classifier Architecture Searchcited by 902
  8. 2017Deep Bayesian Active Learning with Image Datacited by 812
  9. 2018Learning to Reweight Examples for Robust Deep Learningcited by 670
  10. 2018ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardwarecited by 666
  11. 2017Deep Mutual Learningcited by 621
  12. 2017Countering Adversarial Images using Input Transformationscited by 517
  13. 2018FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Searchcited by 480
  14. 2017Regularizing Neural Networks by Penalizing Confident Output Distributionscited by 461
  15. 2019Single Path One-Shot Neural Architecture Search with Uniform Samplingcited by 411
  16. 2016Entropy-SGD: Biasing Gradient Descent Into Wide Valleyscited by 357
  17. 2018Not All Samples Are Created Equal: Deep Learning with Importance Samplingcited by 323
  18. 2019PyHessian: Neural Networks Through the Lens of the Hessiancited by 291
  19. 2018SNAS: Stochastic Neural Architecture Searchcited by 288
  20. 2016Designing Neural Network Architectures using Reinforcement Learningcited by 277
  21. 2019PC-DARTS: Partial Channel Connections for Memory-Efficient Architecture Searchcited by 265
  22. 2018DropBlock: A regularization method for convolutional networkscited by 261
  23. 2020Puzzle Mix: Exploiting Saliency and Local Statistics for Optimal Mixupcited by 260
  24. 2016Deep Pyramidal Residual Networkscited by 163
  25. 2020SaliencyMix: A Saliency Guided Data Augmentation Strategy for Better Regularizationcited by 149
  26. 2021Co-Mixup: Saliency Guided Joint Mixup with Supermodular Diversitycited by 127
  27. 2021Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labelscited by 117
  28. 2019BayesNAS: A Bayesian Approach for Neural Architecture Searchcited by 67
  29. 2020CPR: Classifier-Projection Regularization for Continual Learningcited by 62
  30. 2021Neural Architecture Search with Random Labelscited by 33
  31. 2017Biased Importance Sampling for Deep Neural Network Trainingcited by 31
  32. 2018BlockQNN: Efficient Block-wise Neural Network Architecture Generationcited by 31
  33. 2020An Empirical Evaluation on Robustness and Uncertainty of Regularization Methodscited by 28

Built on by

Cited by year

Citing papers per year: 2022: 2; 2023: 1; 2024: 1.

4 papers.

  1. 2022A Survey of Mix-based Data Augmentation: Taxonomy, Methods, Applications, and Explainabilitycited by 30
  2. 2022A Unified Analysis of Mixed Sample Data Augmentation: A Loss Function Perspectivecited by 27
  3. 2024A Survey on Mixup Augmentations and Beyondcited by 18
  4. 2023ProxyDet: Synthesizing Proxy Novel Classes via Classwise Mixup for Open-Vocabulary Object Detectioncited by 15