# Un-Mix: Rethinking Image Mixtures for Unsupervised Visual Representation Learning

arXiv: https://arxiv.org/abs/2003.05438 · HTML: https://mlsanity.com/paper/2003.05438/

## Key facts

- Cited by 62 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Mar 2020.

## Built on by (most cited first, 30 of 62)

- [Representational Continuity for Unsupervised Continual Learning](https://mlsanity.com/paper/2110.06976/) (2021)
- [TransMix: Attend to Mix for Vision Transformers](https://mlsanity.com/paper/2111.09833/) (2021)
- [Crafting Better Contrastive Views for Siamese Representation Learning](https://mlsanity.com/paper/2202.03278/) (2022)
- [Is Label Smoothing Truly Incompatible with Knowledge Distillation: An Empirical Study](https://mlsanity.com/paper/2104.00676/) (2021)
- [PASS: An ImageNet replacement for self-supervised pretraining without humans](https://mlsanity.com/paper/2109.13228/) (2021)
- [XCon: Learning with Experts for Fine-grained Category Discovery](https://mlsanity.com/paper/2208.01898/) (2022)
- [Contrast and Mix: Temporal Contrastive Video Domain Adaptation with Background Mixing](https://mlsanity.com/paper/2110.15128/) (2021)
- [Improving Contrastive Learning by Visualizing Feature Transformation](https://mlsanity.com/paper/2108.02982/) (2021)
- [Mixed Autoencoder for Self-supervised Visual Representation Learning](https://mlsanity.com/paper/2303.17152/) (2023)
- [Revisiting Label Smoothing and Knowledge Distillation Compatibility: What was Missing?](https://mlsanity.com/paper/2206.14532/) (2022)
- [A Survey of Mix-based Data Augmentation: Taxonomy, Methods, Applications, and Explainability](https://mlsanity.com/paper/2212.10888/) (2022)
- [Geodesic Multi-Modal Mixup for Robust Fine-Tuning](https://mlsanity.com/paper/2203.03897/) (2022)
- [A Simple Data Mixing Prior for Improving Self-Supervised Learning](https://mlsanity.com/paper/2206.07692/) (2022)
- [A Unified Analysis of Mixed Sample Data Augmentation: A Loss Function Perspective](https://mlsanity.com/paper/2208.09913/) (2022)
- [Survey: Image Mixing and Deleting for Data Augmentation](https://mlsanity.com/paper/2106.07085/) (2021)
- [HaLP: Hallucinating Latent Positives for Skeleton-based Self-Supervised Learning of Actions](https://mlsanity.com/paper/2304.00387/) (2023)
- [Boosting Discriminative Visual Representation Learning with Scenario-Agnostic Mixup](https://mlsanity.com/paper/2111.15454/) (2021)
- [OpenMixup: Open Mixup Toolbox and Benchmark for Visual Representation Learning](https://mlsanity.com/paper/2209.04851/) (2022)
- [A Survey on Mixup Augmentations and Beyond](https://mlsanity.com/paper/2409.05202/) (2024)
- [PartMix: Regularization Strategy to Learn Part Discovery for Visible-Infrared Person Re-identification](https://mlsanity.com/paper/2304.01537/) (2023)
- [Prompted Contrast with Masked Motion Modeling: Towards Versatile 3D Action Representation Learning](https://mlsanity.com/paper/2308.03975/) (2023)
- [RankMixup: Ranking-Based Mixup Training for Network Calibration](https://mlsanity.com/paper/2308.11990/) (2023)
- [Finding Order in Chaos: A Novel Data Augmentation Method for Time Series in Contrastive Learning](https://mlsanity.com/paper/2309.13439/) (2023)
- [Harnessing Hard Mixed Samples with Decoupled Regularizer](https://mlsanity.com/paper/2203.10761/) (2022)
- [Contrastive Learning from Spatio-Temporal Mixed Skeleton Sequences for Self-Supervised Skeleton-Based Action Recognition](https://mlsanity.com/paper/2207.03065/) (2022)
- [Hallucination Improves the Performance of Unsupervised Visual Representation Learning](https://mlsanity.com/paper/2307.12168/) (2023)
- [S2-BNN: Bridging the Gap Between Self-Supervised Real and 1-bit Neural Networks via Guided Distribution Calibration](https://mlsanity.com/paper/2102.08946/) (2021)
- [A Random CNN Sees Objects: One Inductive Bias of CNN and Its Applications](https://mlsanity.com/paper/2106.09259/) (2021)
- [Learning Instance-Specific Augmentations by Capturing Local Invariances](https://mlsanity.com/paper/2206.00051/) (2022)
- [Deep Bregman Divergence for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2109.07455/) (2021)

## Abstract (arXiv)

The recently advanced unsupervised learning approaches use the siamese-like framework to compare two "views" from the same image for learning representations. Making the two views distinctive is a core to guarantee that unsupervised methods can learn meaningful information. However, such frameworks are sometimes fragile on overfitting if the augmentations used for generating two views are not strong enough, causing the over-confident issue on the training data. This drawback hinders the model from learning subtle variance and fine-grained information. To address this, in this work we aim to involve the distance concept on label space in the unsupervised learning and let the model be aware of the soft degree of similarity between positive or negative pairs through mixing the input data space, to further work collaboratively for the input and loss spaces. Despite its conceptual simplicity, we show empirically that with the solution -- Unsupervised image mixtures (Un-Mix), we can learn subtler, more robust and generalized representations from the transformed input and corresponding new label space. Extensive experiments are conducted on CIFAR-10, CIFAR-100, STL-10, Tiny ImageNet and standard ImageNet with popular unsupervised methods SimCLR, BYOL, MoCo V1&V2, SwAV, etc. Our proposed image mixture and label assignment strategy can obtain consistent improvement by 1~3% following exactly the same hyperparameters and training procedures of the base methods. Code is publicly available at https://github.com/szq0214/Un-Mix.