Sign in

Efficient Residue Number System Based Winograd Convolution

By Zhi-Gang Liu and Matthew Mattina
Prior research has shown that Winograd algorithm can reduce the computational complexity of convolutional neural networks (CNN) with weights and activations represented in floating point. However it is difficult to apply the scheme to the inference of low-precision quantized (e.g. INT8) networks. Our work extends the Winograd algorithm to Residue... Show more
July 23, 2020
=
0
Loading PDF…
Loading full text...
Similar articles
Loading recommendations...
=
0
x1
Efficient Residue Number System Based Winograd Convolution
Click on play to start listening