FWUA : A Flexible Winograd-Based Uniform Accelerator for 1D/2D/3D CNNs

Wang, Jian; Deng, Huipeng; Ye, Huafeng; Xiao, Shanlin; Yu, Zhiyi

doi:10.1109/icta53157.2021.9661647

Search citation statements

Order By: Relevance

Paper Sections

Select...

Dimension-level Processing Engine Dimension Fusion Is a Coll...1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2021

2024

Publication Types

Select...

Article3

Relationship

Self Cite0

Independent3

Authors

Journals

Cited by 3 publications

(1 citation statement)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fig. 5 shows the arrangement of the transformation matrices [23]. Three paths represent three dimensional operations.…”

Section: Dimension-level Processing Engine Dimension Fusion Is a Coll...mentioning

confidence: 99%

Dimension fusion: Dimension-level dynamically composable accelerator for convolutional neural networks

Deng

Wang

et al. 2021

IEICE Electron. Express

View full text Add to dashboard Cite

Convolutional neural networks (CNNs) have proven to be promising in various applications such as audio recognition, image classification, and video understanding. Winograd algorithm helps to reduce the complexity of computation in a convolution but suffers from poor compatibility for different convolution shapes. This work introduces a dynamic dimension-level fusion architecture based on Winograd for accelerating different dimensions of CNNs. We explore this Winograd architecture by designing Dimension Fusion, a dimension-level processing engine that dynamically fuses to match the convolution shape of individual CNN layers. The proposed architecture is the first work based on Winograd algorithm to be compatible with all convolution shapes (dimension, stride, and filtersize) and achieves highest PE efficiency up to 1.55x and energy efficiency up to 3.3x compared with the state-of-art accelerators.

show abstract

“…Fig. 5 shows the arrangement of the transformation matrices [23]. Three paths represent three dimensional operations.…”

Section: Dimension-level Processing Engine Dimension Fusion Is a Coll...mentioning

confidence: 99%

Dimension fusion: Dimension-level dynamically composable accelerator for convolutional neural networks

Deng

Wang

et al. 2021

IEICE Electron. Express

View full text Add to dashboard Cite

show abstract

Flexible and Efficient Convolutional Acceleration on Unified Hardware Using the Two-Stage Splitting Method and Layer-Adaptive Allocation of 1-D/2-D Winograd Units

Yang,

Meng

et al. 2024

IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst.

View full text Add to dashboard Cite

WRA-SS: A High-Performance Accelerator Integrating Winograd With Structured Sparsity for Convolutional Neural Networks

Yang,

Meng,

et al. 2024

IEEE Trans. VLSI Syst.

View full text Add to dashboard Cite

FWUA : A Flexible Winograd-Based Uniform Accelerator for 1D/2D/3D CNNs

Cited by 3 publications

References 4 publications

Dimension fusion: Dimension-level dynamically composable accelerator for convolutional neural networks

Dimension fusion: Dimension-level dynamically composable accelerator for convolutional neural networks

Flexible and Efficient Convolutional Acceleration on Unified Hardware Using the Two-Stage Splitting Method and Layer-Adaptive Allocation of 1-D/2-D Winograd Units

WRA-SS: A High-Performance Accelerator Integrating Winograd With Structured Sparsity for Convolutional Neural Networks

Contact Info

Product

Resources

About