# Improving Fractal Pre-training

arXiv: https://arxiv.org/abs/2110.03091 · HTML: https://mlsanity.com/paper/2110.03091/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- Builds on 32 of its 50 references that are in the map.
- First posted on arXiv 6 Oct 2021.

## Builds on (most cited first, 30 of 32)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [Fine-Grained Visual Classification of Aircraft](https://mlsanity.com/paper/1306.5151/) (2013)
- [Context Encoders: Feature Learning by Inpainting](https://mlsanity.com/paper/1604.07379/) (2016)
- [Unsupervised Learning of Visual Representations by Solving Jigsaw Puzzles](https://mlsanity.com/paper/1603.09246/) (2016)
- [Colorful Image Colorization](https://mlsanity.com/paper/1603.08511/) (2016)
- [Unsupervised Visual Representation Learning by Context Prediction](https://mlsanity.com/paper/1505.05192/) (2015)
- [Revisiting Unreasonable Effectiveness of Data in Deep Learning Era](https://mlsanity.com/paper/1707.02968/) (2017)
- [Self-supervised Visual Feature Learning with Deep Neural Networks: A Survey](https://mlsanity.com/paper/1902.06162/) (2019)
- [Exploring the Limits of Weakly Supervised Pretraining](https://mlsanity.com/paper/1805.00932/) (2018)
- [Do Better ImageNet Models Transfer Better?](https://mlsanity.com/paper/1805.08974/) (2018)
- [A Survey on Contrastive Self-supervised Learning](https://mlsanity.com/paper/2011.00362/) (2020)
- [Rethinking ImageNet Pre-training](https://mlsanity.com/paper/1811.08883/) (2018)
- [Rethinking Pre-training and Self-training](https://mlsanity.com/paper/2006.06882/) (2020)
- [Training Deep Networks with Synthetic Data: Bridging the Reality Gap by Domain Randomization](https://mlsanity.com/paper/1804.06516/) (2018)
- [Large image datasets: A pyrrhic win for computer vision?](https://mlsanity.com/paper/2006.16923/) (2020)
- [FractalNet: Ultra-Deep Neural Networks without Residuals](https://mlsanity.com/paper/1605.07648/) (2016)
- [Gland Segmentation in Colon Histology Images: The GlaS Challenge Contest](https://mlsanity.com/paper/1603.00275/) (2016)
- [What makes ImageNet good for transfer learning?](https://mlsanity.com/paper/1608.08614/) (2016)
- [Large Scale Fine-Grained Categorization and Domain-Specific Transfer Learning](https://mlsanity.com/paper/1806.06193/) (2018)
- [Synthetic Data for Deep Learning](https://mlsanity.com/paper/1909.11512/) (2019)
- [GAN Augmentation: Augmenting Training Data using Generative Adversarial Networks](https://mlsanity.com/paper/1810.10863/) (2018)
- [Image Representations Learned With Unsupervised Pre-Training Contain Human-like Biases](https://mlsanity.com/paper/2010.15052/) (2020)
- [Pre-training without Natural Images](https://mlsanity.com/paper/2101.08515/) (2021)
- [The Unreasonable Effectiveness of Noisy Data for Fine-Grained Recognition](https://mlsanity.com/paper/1511.06789/) (2015)

## Built on by (most cited first, 26 of 26)

- [A Survey of Deep Learning for Mathematical Reasoning](https://mlsanity.com/paper/2212.10535/) (2022)
- [On the Importance and Applicability of Pre-Training for Federated Learning](https://mlsanity.com/paper/2206.11488/) (2022)
- [A Vision Check-up for Language Models](https://mlsanity.com/paper/2401.01862/) (2024)
- [Asynchronous Federated Continual Learning](https://mlsanity.com/paper/2304.03626/) (2023)
- [Procedural Image Programs for Representation Learning](https://mlsanity.com/paper/2211.16412/) (2022)
- [Insights into Pre-training via Simpler Synthetic Tasks](https://mlsanity.com/paper/2206.10139/) (2022)
- [Pre-training Vision Transformers with Formula-driven Supervised Learning](https://mlsanity.com/paper/2206.09132/) (2022)
- [Visual Atoms: Pre-training Vision Transformers with Sinusoidal Waves](https://mlsanity.com/paper/2303.01112/) (2023)
- [IPMix: Label-Preserving Data Augmentation Method for Training Robust Classifiers](https://mlsanity.com/paper/2310.04780/) (2023)
- [Scaling Backwards: Minimal Synthetic Pre-training?](https://mlsanity.com/paper/2408.00677/) (2024)
- [Pre-training Vision Transformers with Very Limited Synthesized Images](https://mlsanity.com/paper/2307.14710/) (2023)
- [Pre-training with Synthetic Data Helps Offline Reinforcement Learning](https://mlsanity.com/paper/2310.00771/) (2023)
- [SegRCDB: Semantic Segmentation via Formula-Driven Supervised Learning](https://mlsanity.com/paper/2309.17083/) (2023)
- [Learning Fractals by Gradient Descent](https://mlsanity.com/paper/2303.12722/) (2023)
- [FractalAD: A simple industrial anomaly detection method using fractal anomaly generation and backbone knowledge distillation](https://mlsanity.com/paper/2301.12739/) (2023)
- [Learning Representation for Multitask learning through Self Supervised Auxiliary learning](https://mlsanity.com/paper/2409.16651/) (2024)
- [Pre-training for Action Recognition with Automatically Generated Fractal Datasets](https://mlsanity.com/paper/2411.17584/) (2024)
- [VibrantLeaves: A principled parametric image generator for training deep restoration models](https://mlsanity.com/paper/2504.10201/) (2025)
- [Stylized Structural Patterns for Improved Neural Network Pre-training](https://mlsanity.com/paper/2506.19465/) (2025)
- [Fractals as Pre-training Datasets for Anomaly Detection and Localization](https://mlsanity.com/paper/2405.06980/) (2024)
- [LayerMix: Enhanced Data Augmentation through Fractal Integration for Robust Deep Learning](https://mlsanity.com/paper/2501.04861/) (2025)
- [Formula-Supervised Sound Event Detection: Pre-Training Without Real Data](https://mlsanity.com/paper/2504.04428/) (2025)
- [Separating Knowledge and Perception with Procedural Data](https://mlsanity.com/paper/2508.11697/) (2025)
- [Learning with Mandelbrot and Julia](https://mlsanity.com/paper/2509.00903/) (2025)
- [How to Sample High Quality 3D Fractals for Action Recognition Pre-Training?](https://mlsanity.com/paper/2602.11810/) (2026)
- [Amortized Set Prediction for Inverse IFS Reconstruction from Density Maps](https://mlsanity.com/paper/2608.24175/) (2026)

## Abstract (arXiv)

The deep neural networks used in modern computer vision systems require enormous image datasets to train them. These carefully-curated datasets typically have a million or more images, across a thousand or more distinct categories. The process of creating and curating such a dataset is a monumental undertaking, demanding extensive effort and labelling expense and necessitating careful navigation of technical and social issues such as label accuracy, copyright ownership, and content bias. What if we had a way to harness the power of large image datasets but with few or none of the major issues and concerns currently faced? This paper extends the recent work of Kataoka et. al. (2020), proposing an improved pre-training dataset based on dynamically-generated fractal images. Challenging issues with large-scale image datasets become points of elegance for fractal pre-training: perfect label accuracy at zero cost; no need to store/transmit large image archives; no privacy/demographic bias/concerns of inappropriate content, as no humans are pictured; limitless supply and diversity of images; and the images are free/open-source. Perhaps surprisingly, avoiding these difficulties imposes only a small penalty in performance. Leveraging a newly-proposed pre-training task -- multi-instance prediction -- our experiments demonstrate that fine-tuning a network pre-trained using fractals attains 92.7-98.1% of the accuracy of an ImageNet pre-trained network.