A Survey on Mixup Augmentations and Beyond
read the original abstract
As Deep Neural Networks have achieved thrilling breakthroughs in the past decade, data augmentations have garnered increasing attention as regularization techniques when massive labeled data are unavailable. Among existing augmentations, Mixup and relevant data-mixing methods that convexly combine selected samples and the corresponding labels are widely adopted because they yield high performances by generating data-dependent virtual data while easily migrating to various domains. This survey presents a comprehensive review of foundational mixup methods and their applications. We first elaborate on the training pipeline with mixup augmentations as a unified framework containing modules. A reformulated framework could contain various mixup methods and give intuitive operational procedures. Then, we systematically investigate the applications of mixup augmentations on vision downstream tasks, various data modalities, and some analysis \& theorems of mixup. Meanwhile, we conclude the current status and limitations of mixup research and point out further work for effective and efficient mixup augmentations. This survey can provide researchers with the current state of the art in mixup methods and provide some insights and guidance roles in the mixup arena. An online project with this survey is available at https://github.com/Westlake-AI/Awesome-Mixup.
This paper has not been read by Pith yet.
Forward citations
Cited by 5 Pith papers
-
Data augmented bootstrap: Unifying confidence interval construction by approximate invariance
Data augmented bootstrap constructs confidence intervals via approximately invariant transformations, recovering classical bootstrap, conformal prediction, wild bootstrap, and SymmPI as special cases with interpolated...
-
$S^{2}$-FracMix: Label-Preserving Self-Saliency Mixup Augmentation
S²-FracMix creates label-preserving augmented samples via intra-image self-saliency patch mixing and fractal pattern injection, claiming SOTA results across seven benchmarks in classification, robustness, detection, a...
-
SNR-ST-Mix: Sample-specific Neighborhood Regression Mixup for Augmented Spatial Transcriptomics Imputation with Deep Neural Network
SNR-ST-Mix is a geometry- and expression-aware mixup augmentation that constrains interpolation to k-nearest spatial neighbors and weights by transcriptomic similarity for spatial transcriptomics imputation.
-
Beyond Dark Knowledge: Mixup-Based Distillation for Reliable Predictions
Mixup applied only to the student during KD induces independent linearity acquisition that reduces overconfidence by an order of magnitude while improving accuracy, with calibration transferring separately from accuracy.
-
Enhancing Protein Representation Learning via Manifold Restore Mixing
MRM mixes hidden representations of original and DA-augmented proteins and uses a difficulty scheduler on the beta distribution to produce training samples that balance structural fidelity with diversity.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.