AI Paper Atlas

Research map / Architectures and efficiency / Quantization

DSConv: Efficient Convolution Operator

Marcelo Gennari do Nascimento, Roger Fawcett, Victor Adrian Prisacariu

ICCV 2019 · Architectures and efficiency › Quantization · approach: Mixed-precision quantization · 166 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

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
reduce block size from 256 to 8 for 3-bit weights (ablation)DSConv GoogLeNet · ImageNet val, weight-onlytop-1 accuracy: 5.7 -> 56.8Improved
reduce block size from 256 to 4 for 2-bit weights (ablation)DSConv VGG19 · ImageNet val, weight-onlytop-1 accuracy: 11.3 -> 69.4Improved
reduce block size from 256 to 4 for 3-bit weights (ablation)DSConv ResNet50 · ImageNet val, weight-onlytop-1 accuracy: 44.6 -> 74.7Improved
reduce block size from 256 to 4 for 4-bit weights (ablation)DSConv ResNet50 · ImageNet val, weight-onlytop-1 accuracy: 73.0 -> 75.6Improved
add distillation adaptation with unlabelled data (3/3 bits, B=64) (ablation)DSConv ResNet50 · ImageNet valtop-1 accuracy: 54.4 -> 71.5Improved
add distillation adaptation with unlabelled data (3/3 bits, B=64) (ablation)DSConv GoogLeNet · ImageNet valtop-1 accuracy: 29.3 -> 56.7Improved
add distillation adaptation with unlabelled data (3/3 bits, B=64) (ablation)DSConv ResNet18 · ImageNet valtop-1 accuracy: 30.1 -> 65.8Improved
add distillation adaptation with unlabelled data (4/4 bits, B=128) (ablation)DSConv ResNet50 · ImageNet valtop-1 accuracy: 72.6 -> 74.2Improved
lower weight bit width from 4 to 2 without retraining (ablation)DSConv GoogLeNet · ImageNet val, no datatop-1 accuracy: 63.3 -> 27.6Hurt
lower weight bit width from 4 to 2 without retraining (ablation)DSConv ResNet50 · ImageNet val, no datatop-1 accuracy: 75.1 -> 65.1Hurt
minimise KL divergence instead of L2 norm for the shift values (ablation)DSConv weight quantization · ImageNet valtop-1 accuracyNo clear effect
lower activation and weight bits from 8/8 to 4/4 without adaptation (ablation)DSConv ResNet18 · ImageNet val, no datatop-1 accuracy: 69.7 -> 67.3Hurt

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