【Data】 Domain Generalization and Long-tailed dataset
Domain Generalization and Long-tailed dataset
Long-tailed distribution
- (Train) Unbalanced train dataset -> (Test) Balanced test dataset
- Early Methods
- Re-sampling / Synthesizing
- Feature Space Augmentation for Long-Tailed Data, ECCV20
- Using CAM, the class-specific features and class-generic features are divided.
- The Majority Can Help The Minority, CVPR22
- Using CutMix
- Background -> the majority class // Foreground -> the minority class
- Simple but effective
- Feature Space Augmentation for Long-Tailed Data, ECCV20
- Re-weighting
- Focal loss, Class-balanced loss
- Label distribution-aware Margin Loss, NIPS 2019
- Handle decision boundary.
- Utilize ‘Normed Linear’
- the small number of classes -> large margin in the learning rate annealing stage.
- Two-stage (Backbone VS Classifier)
- Decoupling Representation and Classifier, ICLR20
- Train an entire model using class-balanced loss.
- Train a fully connected model using output normalization.
- Weight Balancing, CVPR22
- Weight decay and MaxNorm
- Simple but impactful
- Decoupling Representation and Classifier, ICLR20
- Representation
- Supervised Contrastive learning, CVPR21
- Two classifiers (1) cross entropy (2) supervised contrastive loss
- Parametric Contrastive learning, ICCV21
- Momentum encoder
- Parametric learnable class centers (balanced cross-entropy loss)
- Supervised Contrastive learning, CVPR21
- Logit Adjustment
- Balanced Softmax, NIPS20
- The balanced softmax adjusts the prediction probability.
- Distribution Alignment, CVPR21
- Balanced Softmax, NIPS20
- Ensemble
- Without performance loss In balanced data (False positive ↑), improve performance.
- Bilateral-Branch Network, CVPR20
- Resampling when training a classifier. (reverse distribution)
- Cross entropy when training representation. (uniform distribution)
- RIDE: Routing Diverse Experts, ICLR21
- Multiple experts
- Distribution-aware diversity loss
- Self-supervised Aggregation of Diverse Experts, NIPS22
- Re-sampling / Synthesizing
-