AI Paper Atlas

Research map / Architectures and efficiency / Quantization

CBP: backpropagation with constraint on weight precision using a pseudo-Lagrange multiplier method

Guhyun Kim, Doo Seok Jeong

NeurIPS 2021 · Architectures and efficiency › Quantization · approach: Low-precision training formats · 0 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 CBP constraint training with the gradually vanishing unconstrained-weight window instead of STE-only post-training (ablation)ResNet-18 · ImageNet validation, binary weightstop-1 accuracy: 64.6% -> 66.6%Improved
Remove the gradually vanishing unconstrained-weight window from CBP (ablation)ResNet-18 · ImageNet validation, binary weightstop-1 accuracy: 66.6% -> 60.2%Hurt
Remove the CBP constraint function and use STE-only post-training (ablation)ResNet-18 · ImageNet validation, binary weightsCFS: 1.19 \times 10^{-3} -> 3.58\times10^{-2}Hurt
Use the gradually vanishing window rather than CBP without the window (ablation)ResNet-18 · ImageNet validation, binary weightsCFS: 1.05\times 10^{-5} -> 1.19 \times 10^{-3}Mixed

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