# EMASAM: a Computationally Efficient Sharpness-Aware Minimization via EMA-Guided Perturbations

arXiv: https://arxiv.org/abs/2608.15105 · HTML: https://mlsanity.com/paper/2608.15105/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 29 of its 41 references that are in the map.
- First posted on arXiv 15 Aug 2026.

## Builds on (most cited first, 29 of 29)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms](https://mlsanity.com/paper/1708.07747/) (2017)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [Wide Residual Networks](https://mlsanity.com/paper/1605.07146/) (2016)
- [Improved Regularization of Convolutional Neural Networks with Cutout](https://mlsanity.com/paper/1708.04552/) (2017)
- [Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour](https://mlsanity.com/paper/1706.02677/) (2017)
- [Sharpness-Aware Minimization for Efficiently Improving Generalization](https://mlsanity.com/paper/2010.01412/) (2020)
- [On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima](https://mlsanity.com/paper/1609.04836/) (2016)
- [Computing Nonvacuous Generalization Bounds for Deep (Stochastic) Neural Networks with Many More Parameters than Training Data](https://mlsanity.com/paper/1703.11008/) (2017)
- [Entropy-SGD: Biasing Gradient Descent Into Wide Valleys](https://mlsanity.com/paper/1611.01838/) (2016)
- [A PAC-Bayesian Approach to Spectrally-Normalized Margin Bounds for Neural Networks](https://mlsanity.com/paper/1707.09564/) (2017)
- [ASAM: Adaptive Sharpness-Aware Minimization for Scale-Invariant Learning of Deep Neural Networks](https://mlsanity.com/paper/2102.11600/) (2021)
- [Surrogate Gap Minimization Improves Sharpness-Aware Training](https://mlsanity.com/paper/2203.08065/) (2022)
- [Deep Pyramidal Residual Networks](https://mlsanity.com/paper/1610.02915/) (2016)
- [The Road Less Scheduled](https://mlsanity.com/paper/2405.15682/) (2024)
- [Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks](https://mlsanity.com/paper/1907.04595/) (2019)
- [Stochastic gradient descent performs variational inference, converges to limit cycles for deep networks](https://mlsanity.com/paper/1710.11029/) (2017)
- [Fisher-Rao Metric, Geometry, and Complexity of Neural Networks](https://mlsanity.com/paper/1711.01530/) (2017)
- [Closing the Generalization Gap of Adaptive Gradient Methods in Training Deep Neural Networks](https://mlsanity.com/paper/1806.06763/) (2018)
- [Friendly Sharpness-Aware Minimization](https://mlsanity.com/paper/2403.12350/) (2024)
- [An Adaptive Policy to Employ Sharpness-Aware Minimization](https://mlsanity.com/paper/2304.14647/) (2023)
- [Momentum-SAM: Sharpness Aware Minimization without Computational Overhead](https://mlsanity.com/paper/2401.12033/) (2024)
- [GA-SAM: Gradient-Strength based Adaptive Sharpness-Aware Minimization for Improved Generalization](https://mlsanity.com/paper/2210.06895/) (2022)
- [Breaking Inter-Layer Co-Adaptation by Classifier Anonymization](https://mlsanity.com/paper/1906.01150/) (2019)
- [PoF: Post-Training of Feature Extractor for Improving Generalization](https://mlsanity.com/paper/2207.01847/) (2022)
- [Adaptive Adversarial Cross-Entropy Loss for Sharpness-Aware Minimization](https://mlsanity.com/paper/2406.14329/) (2024)

## Abstract (arXiv)

Recent progress in optimization research has highlighted the sharpness of the loss landscape as a key factor in narrowing the generalization gap. Motivated by this insight, Sharpness-Aware Minimization (SAM) was proposed as a training strategy that enhances generalization. Despite the promising performance, SAM suffers from its twice computational cost due to its core algorithm requiring an extra gradient computation during the perturbation step. To overcome this limitation, we introduce Exponential Moving Average Sharpness-Aware Minimization (EMASAM), a computationally efficient variant of SAM. EMASAM does not require the loss gradient in the perturbation step. Instead, EMASAM defines the perturbation direction based on the discrepancy between the main model and the EMA shadow model. This perturbation travels away from the stable average position toward the less stable area, acting as a softer yet cheaper alternative to SAM's worst-case scenario perturbation. Moreover, since EMASAM's perturbation does not rely on noisy mini-batch gradients, it mitigates the gradient-induced instability inherent in SAM. Hence, EMASAM eliminates the need for an extra backpropagation while also preserving the generalization ability of the SAM-style training. Several experiments have been performed and confirm the efficiency and robustness of our method.