AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Recover Fair Deep Classification Models via Altering Pre-trained Structure

Yanfu Zhang, Shangqian Gao, Heng Huang

ECCV 2022 · Architectures and efficiency › Pruning and efficient tuning · approach: Low-rank compression and merging · 5 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
prune corrupted weights then adversarially fine-tune instead of fine-tuning the full model (ablation)Adv.Ft · CIFAR-10Saccuracy: 0.917 -> 0.920Improved
prune corrupted weights then adversarially fine-tune instead of fine-tuning the full model (ablation)Adv.Ft · CIFAR-10Sbias: 0.051 -> 0.033Improved
graft external filters into corrupted weights instead of noise (ablation)Grafting · CIFAR-10Saccuracy: 0.909 -> 0.927Improved
graft internal filters instead of noise (ablation)Grafting · CIFAR-10Sbias: 0.045 -> 0.013Improved
use network augmentation instead of external graftingExtGraft · CIFAR-10Saccuracy: 0.927 -> 0.931Improved
use network augmentation instead of external graftingExtGraft · CIFAR-10Sbias: 0.028 -> 0.014Improved
use grafting or NetAug instead of prune + adversarial fine-tune on tabular dataPrune + AdvFt · Adult/COMPASaccuracy and biasNo clear effect
use NetAug at bias level 99% instead of Adv.FtAdv.Ft · CIFAR-10S (bias level 99%)accuracy: 0.916 -> 0.915No clear effect
use ExtGraft instead of Adv.Ft at bias level 90%Adv.Ft · CIFAR-10S (bias level 90%)accuracy: 0.937 -> 0.945Improved
use NetAug on CelebA instead of Adv.FtAdv.Ft · CelebAaccuracy: 0.91 -> 0.94Improved
prune unfair model guided by ExtGraft teacher at 99% imbalance (ablation)Unfair full model · CIFAR-10Sbias: 0.115 -> 0.082Improved

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