# PatchShuffle Regularization

arXiv: https://arxiv.org/abs/1707.07103 · HTML: https://mlsanity.com/paper/1707.07103/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Jul 2017.

## Built on by (most cited first, 23 of 23)

- [A Comprehensive Survey on Data Augmentation](https://mlsanity.com/paper/2405.09591/) (2024)
- [TransMEF: A Transformer-Based Multi-Exposure Image Fusion Framework using Self-Supervised Multi-Task Learning](https://mlsanity.com/paper/2112.01030/) (2021)
- [Toward Understanding Generative Data Augmentation](https://mlsanity.com/paper/2305.17476/) (2023)
- [A Survey on the Robustness of Computer Vision Models against Common Corruptions](https://mlsanity.com/paper/2305.06024/) (2023)
- [Sill-Net: Feature Augmentation with Separated Illumination Representation](https://mlsanity.com/paper/2102.03539/) (2021)
- [Medical Image Segmentation via Single-Source Domain Generalization with Random Amplitude Spectrum Synthesis](https://mlsanity.com/paper/2409.04768/) (2024)
- [On-Device Learning with Cloud-Coordinated Data Augmentation for Extreme Model Personalization in Recommender Systems](https://mlsanity.com/paper/2201.10382/) (2022)
- [Enhancing Sample Utilization through Sample Adaptive Augmentation in Semi-Supervised Learning](https://mlsanity.com/paper/2309.03598/) (2023)
- [Block shuffling learning for Deepfake Detection](https://mlsanity.com/paper/2202.02819/) (2022)
- [Shuffled Patch-Wise Supervision for Presentation Attack Detection](https://mlsanity.com/paper/2109.03484/) (2021)
- [Ortho-Shot: Low Displacement Rank Regularization with Data Augmentation for Few-Shot Learning](https://mlsanity.com/paper/2110.09374/) (2021)
- [Feature transforms for image data augmentation](https://mlsanity.com/paper/2201.09700/) (2022)
- [Meet-in-the-middle: Multi-scale upsampling and matching for cross-resolution face recognition](https://mlsanity.com/paper/2211.15225/) (2022)
- [A Comprehensive Methodological Survey of Human Activity Recognition Across Divers Data Modalities](https://mlsanity.com/paper/2409.09678/) (2024)
- [Random Padding Data Augmentation](https://mlsanity.com/paper/2302.08682/) (2023)
- [BadVim: Unveiling Backdoor Threats in Visual State Space Model](https://mlsanity.com/paper/2408.11679/) (2024)
- [ShuffleBlock: Shuffle to Regularize Deep Convolutional Neural Networks](https://mlsanity.com/paper/2106.09358/) (2021)
- [Redistribute Ensemble Training for Mitigating Memorization in Diffusion Models](https://mlsanity.com/paper/2502.09434/) (2025)
- [Relevance-driven Input Dropout: an Explanation-guided Regularization Technique](https://mlsanity.com/paper/2505.21595/) (2025)
- [Anatomy-Aware Lymphoma Lesion Detection in Whole-Body PET/CT](https://mlsanity.com/paper/2511.07047/) (2025)
- [Sliding-Window Reordering with Overlap Averaging: A Simple Time-Domain Augmentation for Multivariate Forecasting](https://mlsanity.com/paper/2604.09067/) (2026)
- [Reconstructing Randomly Masked Spectra Helps DNNs Identify Discriminant Wavenumbers](https://mlsanity.com/paper/2606.21289/) (2026)
- [ICM: Intra-class Mixing for Domain Adaptation in Adverse Weather](https://mlsanity.com/paper/2609.27533/) (2026)

## Abstract (arXiv)

This paper focuses on regularizing the training of the convolutional neural network (CNN). We propose a new regularization approach named ``PatchShuffle`` that can be adopted in any classification-oriented CNN models. It is easy to implement: in each mini-batch, images or feature maps are randomly chosen to undergo a transformation such that pixels within each local patch are shuffled. Through generating images and feature maps with interior orderless patches, PatchShuffle creates rich local variations, reduces the risk of network overfitting, and can be viewed as a beneficial supplement to various kinds of training regularization techniques, such as weight decay, model ensemble and dropout. Experiments on four representative classification datasets show that PatchShuffle improves the generalization ability of CNN especially when the data is scarce. Moreover, we empirically illustrate that CNN models trained with PatchShuffle are more robust to noise and local changes in an image.