A Reconfigurable Winograd CNN Accelerator with Nesting Decomposition Algorithm for Computing Convolution with Large Filters

02/26/2021
by   Jingbo Jiang, et al.
0

Recent literature found that convolutional neural networks (CNN) with large filters perform well in some applications such as image semantic segmentation. Winograd transformation helps to reduce the number of multiplications in a convolution but suffers from numerical instability when the convolution filter size gets large. This work proposes a nested Winograd algorithm to iteratively decompose a large filter into a sequence of 3x3 tiles which can then be accelerated with a 3x3 Winograd algorithm. Compared with the state-of-art OLA-Winograd algorithm, the proposed algorithm reduces the multiplications by 1.41 to 3.29 times for computing 5x5 to 9x9 convolutions.

READ FULL TEXT
research
01/05/2018

Learning 3D-FilterMap for Deep Convolutional Neural Networks

We present a novel and compact architecture for deep Convolutional Neura...
research
07/23/2020

Efficient Residue Number System Based Winograd Convolution

Prior research has shown that Winograd algorithm can reduce the computat...
research
07/26/2017

Optimizing Filter Size in Convolutional Neural Networks for Facial Action Unit Recognition

Recognizing facial action units (AUs) during spontaneous facial displays...
research
08/28/2018

Convolutional Neural Networks with Recurrent Neural Filters

We introduce a class of convolutional neural networks (CNNs) that utiliz...
research
09/25/2020

Tied Block Convolution: Leaner and Better CNNs with Shared Thinner Filters

Convolution is the main building block of convolutional neural networks ...
research
05/02/2022

Efficient Accelerator for Dilated and Transposed Convolution with Decomposition

Hardware acceleration for dilated and transposed convolution enables rea...
research
06/15/2018

Learning Front-end Filter-bank Parameters using Convolutional Neural Networks for Abnormal Heart Sound Detection

Automatic heart sound abnormality detection can play a vital role in the...

Please sign up or login with your details

Forgot password? Click here to reset