AI Paper Atlas

Research map / Adversarial robustness and security / Backdoor and poisoning

One-shot Neural Backdoor Erasing via Adversarial Weight Masking

Shuwen Chai, Jinghui Chen

NeurIPS 2022 · Adversarial robustness and security › Backdoor and poisoning · approach: Training-time backdoor defense · 3 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
replace neuron pruning with soft weight masking and recover adversarial triggersbackdoor removal on CIFAR-10 · CIFAR-10 one-shot, BadNetsASR: 32.83 -> 8.98Improved
replace neuron pruning with soft weight masking and recover adversarial triggersbackdoor removal on CIFAR-10 · CIFAR-10 one-shot, BadNetsACC: 60.35 -> 76.46Improved
add soft weight masking to ANP (ablation)ANP · CIFAR-10 one-shot, BadNetsACC: 53.26 -> 73.34Improved
add adversarial trigger recovery to ANP + SWM (ablation)ANP + SWM · CIFAR-10 one-shot, BadNetsACC: 73.34 -> 80.38Improved
add adversarial trigger recovery to ANP + SWM (ablation)ANP + SWM · CIFAR-10 one-shot, BadNetsASR: 10.16 -> 10.41No clear effect
add L1 mask regularization to ANP + SWM + ATR (ablation)ANP + SWM + ATR · CIFAR-10 one-shot, BadNetsASR: 10.41 -> 8.98Improved
remove clipping of the recovered perturbation (ablation)AWM · CIFAR-10 one-shot, BadNetsASR: 8.98 -> 16.28Hurt
remove clipping of the recovered perturbation (ablation)AWM · CIFAR-10 one-shot, BadNetsACC: 76.46 -> 66.03Hurt
remove L1 regularization on the mask (ablation)AWM · CIFAR-10 one-shot, BadNetsACC: 76.46 -> 66.32Hurt
remove perturbation clipping and mask L1 regularization (ablation)AWM · CIFAR-10 one-shot, BadNetsACC: 76.46 -> 65.32Hurt
replace the L1 perturbation constraint with an L2 constraint (ablation)AWM · CIFAR-10 one-shot, BadNetsASR: 8.98 -> 8.98No clear effect
use an L2 perturbation constraint and remove perturbation clipping (ablation)AWM · CIFAR-10 one-shot, BadNetsASR: 8.98 -> 14.38Hurt

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