AI Paper Atlas

Research map / Architectures and efficiency / Quantization

PalQuant: Accelerating High-Precision Networks on Low-Precision Accelerators

Qinghao Hu, Gang Li, Qiman Wu, Jian Cheng

ECCV 2022 · Architectures and efficiency › Quantization · approach: Low-precision training formats · 2 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
add cyclic shuffle and per-stage channel shuffle (ablation)PalQuant B=2,G=2 · ImageNet ResNet-18top-1 accuracy: 67.94 -> 71.13Improved
add per-block channel shuffle alone (ablation)PalQuant B=2,G=2 · ImageNet ResNet-18top-1 accuracy: 67.94 -> 70.31Improved
add cyclic shuffle alone (ablation)PalQuant B=2,G=2 · ImageNet ResNet-18top-1 accuracy: 67.94 -> 70.22Improved
add cyclic shuffle to per-stage channel shuffle (ablation)PalQuant B=2,G=2 · ImageNet ResNet-18top-1 accuracy: 70.24 -> 71.13Improved
use per-stage instead of per-block channel shuffle with cyclic shuffle (ablation)PalQuant B=2,G=2 · ImageNet ResNet-18top-1 accuracy: 70.62 -> 71.13Improved
add cyclic permutation inside the cyclic shuffle (ablation)PalQuant B=2,G=2 · ImageNet ResNet-18top-1 accuracy: 70.37 -> 71.13Improved
add the whole cyclic shuffle module (ablation)PalQuant B=2,G=2 · ImageNet ResNet-18top-1 accuracy: 70.23 -> 71.13Improved
add cyclic shuffle to ShuffleNet-v2ShuffleNet v2 · ImageNettop-1 accuracy: 68.71 -> 69.19Improved
parallel low-precision groups (G=2) instead of LSQ 4bPalQuant vs LSQ · ImageNet ResNet-18top-1 accuracy: 70.70 -> 71.12Improved
increase groups G from 2 to 4 (ablation)PalQuant B=2 · ImageNet ResNet-18top-1 accuracy: 71.12 -> 72.74Improved
PalQuant instead of LSQ at matched BitOps (6b A,2b W)PalQuant B=2,G=3 · ImageNet ResNet-18top-1 accuracy: 69.47 -> 72.36Improved
increase B from 2 to 4 with G=2 (ablation)PalQuant · ImageNet ResNet-18top-1 accuracy: 71.13 -> 73.48Improved

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