# Data Augmentation with Variational Autoencoders and Manifold Sampling

arXiv: https://arxiv.org/abs/2103.13751 · HTML: https://mlsanity.com/paper/2103.13751/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 22 references that are in the map.
- First posted on arXiv 25 Mar 2021.

## Builds on (most cited first, 8 of 8)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms](https://mlsanity.com/paper/1708.07747/) (2017)
- [Data Augmentation Generative Adversarial Networks](https://mlsanity.com/paper/1711.04340/) (2017)
- [VAE with a VampPrior](https://mlsanity.com/paper/1705.07120/) (2017)
- [Latent Space Oddity: on the Curvature of Deep Generative Models](https://mlsanity.com/paper/1710.11379/) (2017)
- [Data Augmentation in Emotion Classification Using Generative Adversarial Networks](https://mlsanity.com/paper/1711.00648/) (2017)
- [Geometry-Aware Hamiltonian Variational Auto-Encoder](https://mlsanity.com/paper/2010.11518/) (2020)

## Built on by (most cited first, 16 of 16)

- [Deep Learning Approaches for Data Augmentation in Medical Imaging: A Review](https://mlsanity.com/paper/2307.13125/) (2023)
- [Generative AI Enables Medical Image Segmentation in Ultra Low-Data Regimes](https://mlsanity.com/paper/2408.17421/) (2024)
- [RDA-INR: Riemannian Diffeomorphic Autoencoding via Implicit Neural Representations](https://mlsanity.com/paper/2305.12854/) (2023)
- [Anchor Data Augmentation](https://mlsanity.com/paper/2311.06965/) (2023)
- [MGAug: Multimodal Geometric Augmentation in Latent Spaces of Image Deformations](https://mlsanity.com/paper/2312.13440/) (2023)
- [SAVAE: Leveraging the variational Bayes autoencoder for survival analysis](https://mlsanity.com/paper/2312.14651/) (2023)
- [D3A-TS: Denoising-Driven Data Augmentation in Time Series](https://mlsanity.com/paper/2312.05550/) (2023)
- [Discriminative Hamiltonian Variational Autoencoder for Accurate Tumor Segmentation in Data-Scarce Regimes](https://mlsanity.com/paper/2406.11659/) (2024)
- [A Survey on the Integration of Generative AI for Critical Thinking in Mobile Networks](https://mlsanity.com/paper/2404.06946/) (2024)
- [Riemannian Geometry-Preserving Variational Autoencoder for MI-BCI Data Augmentation](https://mlsanity.com/paper/2603.10563/) (2026)
- [Unlocking Latent Dimensions: Exploring Representations of Large-Scale X-ray Scattering Data using Variational Autoencoders](https://mlsanity.com/paper/2606.14999/) (2026)
- [Improving the Security of Smartwatch Payment with Deep Learning](https://mlsanity.com/paper/2307.05437/) (2023)
- [Collinear datasets augmentation using Procrustes validation sets](https://mlsanity.com/paper/2312.04911/) (2023)
- [Evaluating the Stability of Deep Learning Latent Feature Spaces](https://mlsanity.com/paper/2402.11404/) (2024)
- [FAAGC: Feature Augmentation on Adaptive Geodesic Curve Based on the shape space theory](https://mlsanity.com/paper/2501.18619/) (2025)
- [Data-driven Prediction of Species-Specific Plant Responses to Spectral-Shifting Films from Leaf Phenotypic and Photosynthetic Traits](https://mlsanity.com/paper/2511.15173/) (2025)

## Abstract (arXiv)

We propose a new efficient way to sample from a Variational Autoencoder in the challenging low sample size setting. This method reveals particularly well suited to perform data augmentation in such a low data regime and is validated across various standard and real-life data sets. In particular, this scheme allows to greatly improve classification results on the OASIS database where balanced accuracy jumps from 80.7% for a classifier trained with the raw data to 88.6% when trained only with the synthetic data generated by our method. Such results were also observed on 3 standard data sets and with other classifiers. A code is available at https://github.com/clementchadebec/Data_Augmentation_with_VAE-DALI.