AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Defending against Model Stealing via Verifying Embedded External Features

Yiming Li, Linghui Zhu, Xiaojun Jia, Yong Jiang, Shu-Tao Xia, Xiaochun Cao

AAAI 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Data augmentation and generalization bias · 67 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 style transfer to embed external features instead of a white-square patch (ablation)Ownership verification against model stealing · CIFAR-10 / Distillation, Zero-shot, Fine-tuning, Label-query, Logit-queryp-value: 0.17 -> 10^{-7}; 0.01 -> 10^{-5}; 10^{-3} -> 10^{-6}; 10^{-3} -> 10^{-4}; 10^{-3} -> 10^{-4}Improved
Add the meta-classifier to BadNets verification (ablation)BadNets-based model watermarking · CIFAR-10 / Distillation, Zero-shot, Fine-tuning, Label-query, Logit-queryp-value: 0.32 -> 10^{-3}; 0.22 -> 10^{-61}; 0.28 -> 10^{-5}; 0.20 -> 10^{-50}; 0.23 -> 10^{-3}Improved
Use the sign of gradients rather than raw gradients as meta-classifier input (ablation)Ownership meta-classifier · CIFAR-10 and ImageNet / model-stealing attacksp-value: CIFAR-10: 0.01 -> 10^{-7}; ImageNet: 10^{-4} -> 10^{-5}Improved
Increase the transformation rate (ablation)Victim model training and ownership verification · Model-stealing attacksp-valueImproved
Increase the number of sampled transformed images (ablation)Hypothesis-test verification · Ownership verificationp-valueImproved
Change the style image (ablation)External-feature embedding · CIFAR-10 / model-stealing attacksp-valueNo clear effect
Use testing images instead of training images to generate verification samples (ablation)Ownership verification · CIFAR-10 and ImageNet / model-stealing attacksconfidence score (Delta mu)No clear effect
Use style-transformed samples to train the victim modelVictim model · CIFAR-10 / accuracyaccuracy: 91.99\% -> 91.79\%No clear effect
Use BadNets rather than external-feature verificationModel stealing defense · CIFAR-10 / stolen model after zero-shot distillationattack success rate: 98.02 -> 3.84Hurt
Apply dataset inference to unrelated models trained on disjoint subsetsDataset inference · CIFAR-10 / ownership verificationp-value: 10^{-7}; 10^{-5}; 10^{-4}Hurt
Use external-feature verification rather than the listed model-watermarking baselinesOwnership verification · CIFAR-10 and ImageNet / model-stealing attacksp-value: CIFAR-10 distillation: 0.33 -> 10^{-7}; ImageNet distillation: 0.19 -> 10^{-5}Improved

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