# Decorrelation Speeds Up Vision Transformers

arXiv: https://arxiv.org/abs/2510.14657 · HTML: https://mlsanity.com/paper/2510.14657/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 35 references that are in the map.
- First posted on arXiv 16 Oct 2025.

## Builds on (most cited first, 21 of 21)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [Unified Perceptual Parsing for Scene Understanding](https://mlsanity.com/paper/1807.10221/) (2018)
- [Image Transformer](https://mlsanity.com/paper/1802.05751/) (2018)
- [Stand-Alone Self-Attention in Vision Models](https://mlsanity.com/paper/1906.05909/) (2019)
- [On the Relationship between Self-Attention and Convolutional Layers](https://mlsanity.com/paper/1911.03584/) (2019)
- [Estimating the Carbon Footprint of BLOOM, a 176B Parameter Language Model](https://mlsanity.com/paper/2211.02001/) (2022)
- [Power Hungry Processing: Watts Driving the Cost of AI Deployment?](https://mlsanity.com/paper/2311.16863/) (2023)
- [Mixed supervision for surface-defect detection: from weakly to fully supervised learning](https://mlsanity.com/paper/2104.06064/) (2021)
- [Self Pre-training with Masked Autoencoders for Medical Image Classification and Segmentation](https://mlsanity.com/paper/2203.05573/) (2022)
- [Natural Neural Networks](https://mlsanity.com/paper/1507.00210/) (2015)
- [Hi-End-MAE: Hierarchical encoder-driven masked autoencoders are stronger vision learners for medical image segmentation](https://mlsanity.com/paper/2502.08347/) (2025)
- [Constrained Parameter Inference as a Principle for Learning](https://mlsanity.com/paper/2203.13203/) (2022)
- [Efficient Deep Learning with Decorrelated Backpropagation](https://mlsanity.com/paper/2405.02385/) (2024)
- [Correlations Are Ruining Your Gradient Descent](https://mlsanity.com/paper/2407.10780/) (2024)
- [Annotation-cost Minimization for Medical Image Segmentation using Suggestive Mixed Supervision Fully Convolutional Networks](https://mlsanity.com/paper/1812.11302/) (2018)
- [VIS-MAE: An Efficient Self-supervised Learning Approach on Medical Image Segmentation and Classification](https://mlsanity.com/paper/2402.01034/) (2024)

## Abstract (arXiv)

Masked Autoencoder (MAE) pre-training of vision transformers (ViTs) yields strong performance in low-label data regimes but comes with substantial computational costs, making it impractical in time- and resource-constrained industrial settings. We address this by integrating Decorrelated Backpropagation (DBP) into MAE pre-training, an optimization method that iteratively reduces input correlations at each layer to accelerate convergence. Applied selectively to the encoder, DBP achieves faster pre-training without loss of stability. To mimic constrained-data scenarios, we evaluate our approach on ImageNet-1K pre-training and ADE20K fine-tuning using randomly sampled subsets of each dataset. Under this setting, DBP-MAE reduces wall-clock time to baseline performance by 21.1%, lowers carbon emissions by 21.4%, and improves segmentation mIoU by 1.1 points. We observe similar gains when pre-training and fine-tuning on proprietary industrial data, confirming the method's applicability in real-world scenarios. These results demonstrate that DBP can reduce training time and energy use while improving downstream performance for large-scale ViT pre-training. Keywords: Deep learning, Vision transformers, Efficient AI, Decorrelation