# DWM: A Decomposable Winograd Method for Convolution Acceleration

arXiv: https://arxiv.org/abs/2002.00552 · HTML: https://mlsanity.com/paper/2002.00552/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Feb 2020.

## Built on by (most cited first, 5 of 5)

- [Winograd Convolution for Deep Neural Networks: Efficient Point Selection](https://mlsanity.com/paper/2201.10369/) (2022)
- [Winograd Convolution: A Perspective from Fault Tolerance](https://mlsanity.com/paper/2202.08675/) (2022)
- [Exploring Winograd Convolution for Cost-effective Neural Network Fault Tolerance](https://mlsanity.com/paper/2308.08230/) (2023)
- [SFC: Achieve Accurate Fast Convolution under Low-precision Arithmetic](https://mlsanity.com/paper/2407.02913/) (2024)
- [Fast Convolution based on Winograd Minimum Filtering: Introduction and Development](https://mlsanity.com/paper/2111.00977/) (2021)

## Abstract (arXiv)

Winograd's minimal filtering algorithm has been widely used in Convolutional Neural Networks (CNNs) to reduce the number of multiplications for faster processing. However, it is only effective on convolutions with kernel size as 3x3 and stride as 1, because it suffers from significantly increased FLOPs and numerical accuracy problem for kernel size larger than 3x3 and fails on convolution with stride larger than 1. In this paper, we propose a novel Decomposable Winograd Method (DWM), which breaks through the limitation of original Winograd's minimal filtering algorithm to a wide and general convolutions. DWM decomposes kernels with large size or large stride to several small kernels with stride as 1 for further applying Winograd method, so that DWM can reduce the number of multiplications while keeping the numerical accuracy. It enables the fast exploring of larger kernel size and larger stride value in CNNs for high performance and accuracy and even the potential for new CNNs. Comparing against the original Winograd, the proposed DWM is able to support all kinds of convolutions with a speedup of ~2, without affecting the numerical accuracy.