AI Paper Atlas

Research map / Architectures and efficiency / Quantization

Towards Mixed-Precision Quantization of Neural Networks via Constrained Optimization

Weihan Chen, Peisong Wang, Jian Cheng

ICCV 2021 · Architectures and efficiency › Quantization · approach: Mixed-precision quantization · 70 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
use second-order term instead of first-order only for loss perturbation (ablation)mixed-precision bit-width assignment · ImageNet, ResNet-18/50, MobileNet-V2accuracyImproved
use Hessian-based loss perturbation instead of Hessian-free 1/2 dw^T dw (ablation)bit-width assignment · ImageNet, deep and lightweight netsaccuracyImproved
add first-order term to second-order term (ablation)loss perturbation criterion · ImageNetaccuracyHurt
use original greedy priority instead of reversed criterion (ablation)MCKP greedy search · ImageNetaccuracyImproved
use original greedy priority instead of random criterion (ablation)MCKP greedy search · ImageNetaccuracyImproved
use mixed precision (Ours) instead of LQ-Nets uniform 4-bit weightsResNet-18 · ImageNetTop-1 drop: -0.30 -> -0.26Improved
use our method instead of HAWQResNet-50 W2MP A4MP · ImageNetTop-1 drop: -1.91 -> -0.85Improved
use our method instead of HAQMobileNet-V2 (13.99x vs 14.07x) · ImageNetTop-1 accuracy: 66.75 -> 68.52Improved
use only 1024 sampled images for loss perturbationbit-width assignment · ImageNetloss perturbation convergenceNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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