Robust adversarial inputs

Robust adversarial inputs | OpenAI

July 17, 2017

Publication

Robust adversarial inputs

Read paper

Loading…

Share

We’ve created images that reliably fool neural network classifiers when viewed from varied scales and perspectives. This challenges a claim from last week that self-driving cars would be hard to trick maliciously since they capture images from multiple scales, angles, perspectives, and the like.

Loading...

Out-of-the-box adversarial examples⁠ do fail under image transformations. Below, we show the same cat picture, adversarially perturbed to be incorrectly classified as a desktop computer by Inception v3⁠ trained on ImageNet⁠. A zoom of as little as 1.002 causes the classification probability for the correct label tabby cat to override the adversarial labeldesktop computer.

Loading...

However, we’d suspected that active effort could produce a robust adversarial example, as adversarial examples have been shown to transfer⁠ to the physical world.

Scale-invariant adversarial examples

Adversarial examples can be created using an optimization method called projected gradient descent to find small perturbations to the image that arbitrarily fool the classifier.

Instead of optimizing for finding an input that’s adversarial from a single viewpoint, we optimize over a large ensemble⁠ of stochastic classifiers that randomly rescale the input before classifying it. Optimizing against such an ensemble produces robust adversarial examples that are scale-invariant.

Loading...

Even when we restrict ourselves to only modifying pixels corresponding to the cat, we can create a single perturbed image that is simultaneously adversarial at all desired scales.

Transformation-invariant adversarial examples

By adding random rotations, translations, scales, noise, and mean shifts to our training perturbations, the same technique produces a single input that remains adversarial under any of these transformations.

Loading...

Our transformations are sampled randomly at test time, demonstrating that our example is invariant to the whole distribution of transformations.

• Ethics & Safety

Author

Anish Athalye

Related articles

View all

Point-E: A system for generating 3D point clouds from complex promptsPublicationDec 16, 2022

Multimodal neurons in artificial neural networksMilestoneMar 4, 2021

CLIP: Connecting text and imagesMilestoneJan 5, 2021