# PatchUp: A Feature-Space Block-Level Regularization Technique for Convolutional Neural Networks

arXiv: https://arxiv.org/abs/2006.07794 · HTML: https://mlsanity.com/paper/2006.07794/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Jun 2020.

## Built on by (most cited first, 12 of 12)

- [A Survey on Mixup Augmentations and Beyond](https://mlsanity.com/paper/2409.05202/) (2024)
- [RandoMix: A mixed sample data augmentation method with multiple mixed modes](https://mlsanity.com/paper/2205.08728/) (2022)
- [MIDAS: Mixing Ambiguous Data with Soft Labels for Dynamic Facial Expression Recognition](https://mlsanity.com/paper/2503.00202/) (2025)
- [CellMix: A General Instance Relationship based Method for Data Augmentation Towards Pathology Image Classification](https://mlsanity.com/paper/2301.11513/) (2023)
- [Diff-Oracle: Deciphering Oracle Bone Scripts with Controllable Diffusion Model](https://mlsanity.com/paper/2312.13631/) (2023)
- [Augmenting Safety-Critical Driving Scenarios while Preserving Similarity to Expert Trajectories](https://mlsanity.com/paper/2404.13347/) (2024)
- [Towards Generalizable Deepfake Detection by Primary Region Regularization](https://mlsanity.com/paper/2307.12534/) (2023)
- [Local and Global Context-and-Object-part-Aware Superpixel-based Data Augmentation for Deep Visual Recognition](https://mlsanity.com/paper/2512.00130/) (2025)
- [Feature Weaken: Vicinal Data Augmentation for Classification](https://mlsanity.com/paper/2211.10944/) (2022)
- [Random Token Fusion for Multi-View Medical Diagnosis](https://mlsanity.com/paper/2410.15847/) (2024)
- [Combining Discrepancy-Confusion Uncertainty and Calibration Diversity for Active Fine-Grained Image Classification](https://mlsanity.com/paper/2509.24181/) (2025)
- [When Geometry Aligns: Dihedral Hidden-State Transformations in UNet, ViT, and DiT Architectures](https://mlsanity.com/paper/2607.03580/) (2026)

## Abstract (arXiv)

Large capacity deep learning models are often prone to a high generalization gap when trained with a limited amount of labeled training data. A recent class of methods to address this problem uses various ways to construct a new training sample by mixing a pair (or more) of training samples. We propose PatchUp, a hidden state block-level regularization technique for Convolutional Neural Networks (CNNs), that is applied on selected contiguous blocks of feature maps from a random pair of samples. Our approach improves the robustness of CNN models against the manifold intrusion problem that may occur in other state-of-the-art mixing approaches. Moreover, since we are mixing the contiguous block of features in the hidden space, which has more dimensions than the input space, we obtain more diverse samples for training towards different dimensions. Our experiments on CIFAR10/100, SVHN, Tiny-ImageNet, and ImageNet using ResNet architectures including PreActResnet18/34, WRN-28-10, ResNet101/152 models show that PatchUp improves upon, or equals, the performance of current state-of-the-art regularizers for CNNs. We also show that PatchUp can provide a better generalization to deformed samples and is more robust against adversarial attacks.