AI Paper Atlas

Research map / Adversarial robustness and security / Certified and empirical robustness

Adversarial Reprogramming of Neural Networks

Gamaleldin F. Elsayed, Ian Goodfellow, Jascha Sohl-Dickstein

ICLR 2019 · Adversarial robustness and security › Certified and empirical robustness · approach: Regularized defenses against attacks · 21 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 a shared learned input program to repurpose ImageNet classifiers for square countingImageNet classification models · Counting squaresaccuracy: 0.9993; 0.9999; 0.9994; 0.9763; 0.9843; 0.9966Improved
Add a shared learned input program to repurpose ImageNet classifiers for digit classificationImageNet classification models · MNIST testaccuracy: 0.9753; 0.9646; 0.9744; 0.9534; 0.9664; 0.9496Improved
Add a shared learned input program to repurpose ImageNet classifiers for object classificationImageNet classification models · CIFAR-10 testaccuracy: 0.6911; 0.6683; 0.6719; 0.6210; 0.6301; 0.5858Improved
Add a learned program for shuffled rather than normally structured MNIST digitsImageNet classification models · Shuffled MNIST testaccuracy: 0.9709; 0.9715; 0.9683; 0.9691; 0.9678; 0.9717Improved
Use adversarially trained rather than standard Inception V3 weights (ablation)Adversarial reprogramming for MNIST · MNIST testaccuracy: 0.9753 -> 0.9752No clear effect
Initialize the target network randomly instead of using ImageNet-trained weights (ablation)Adversarial reprogramming for MNIST · MNIST testaccuracy: 0.9753 -> 0.4539; 0.9646 -> 0.1861; 0.9744 -> 0.1135; 0.9534 -> 0.1032; 0.9664 -> 0.1756; 0.9496 -> 0.9325Hurt
Shuffle the pixels of adversarial MNIST inputs (ablation)Adversarial reprogramming for MNIST · Shuffled MNIST testaccuracy: 0.9709; 0.9715; 0.9683; 0.9691; 0.9678; 0.9717No clear effect
Limit the size of the adversarial program (ablation)Inception V3 reprogrammed for MNIST · MNIST classificationaccuracyHurt
Limit the perturbation scale to make the program nearly imperceptible (ablation)Inception V3 reprogrammed for MNIST · MNIST classificationaccuracyHurt
Hide shuffled MNIST data and the program inside a normal ImageNet image (ablation)Inception V3 reprogrammed for MNIST · MNIST classificationaccuracyHurt

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