AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Efficient Residue Number System Based Winograd Convolution

Zhi-Gang Liu, Matthew Mattina

ECCV 2020 · Architectures and efficiency › CNN design · approach: Compact CNN architectures · 15 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use 16-bit RNS(4001,4331) instead of 8-bit RNS(251,241,239) (ablation)F(14x14,3x3) Winograd · VGG16 INT8 on Arm Cortex-A73average speed-up: 2.02 -> 2.20Improved
use 2-residue 16-bit RNS instead of 3-residue 8-bit RNS (ablation)F(10x10,3x3) Winograd · Arithmetic complexity reductionarithmetic complexity reduction: 2.08 -> 3.13Improved
use 2-residue RNS(4001,4331) instead of 3-residue RNS for large tiles (ablation)F(12x12,5x5) Winograd · Arithmetic complexity reductionarithmetic complexity reduction: 4.69 -> 7.03Improved
increase Winograd tile from 2x2 to 14x14 output (ablation)Winograd over RNS(251,241,239) · Arithmetic complexity reduction (3x3)arithmetic complexity reduction: 0.75 -> 2.30Improved
use 5x5 filters instead of 3x3 filters in RNS Winograd (ablation)F(12x12,k) over RNS(251,241,239) · Arithmetic complexity reductionarithmetic complexity reduction: 2.20 -> 4.69Improved
move from conv1_2 to deeper VGG16 layers (ablation)F(14x14,3x3) RNS(251,241,239) · VGG16 layersspeed-up: 1.86 -> 2.22Improved
apply RNS Winograd to 8-bit CNNs without precision scalingRNS Winograd · Quantized CNN accuracytop-1 accuracyNo clear effect
use larger Winograd tile in standard integer Winograd (no RNS) (ablation)Integer Winograd · Transform data widthtransform data width (bits): 12 -> 18Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy