Explaining and Harnessing Adversarial Examples

Resource | v1 | created by janarez |
Type Paper
Created 2015-03-20
Identifier unavailable


Several machine learning models, including neural networks, consistently misclassify adversarial examples---inputs formed by applying small but intentionally worst-case perturbations to examples from the dataset, such that the perturbed input results in the model outputting an incorrect answer with high confidence. Early attempts at explaining this phenomenon focused on nonlinearity and overfitting. We argue instead that the primary cause of neural networks' vulnerability to adversarial perturbation is their linear nature. This explanation is supported by new quantitative results while giving the first explanation of the most intriguing fact about them: their generalization across architectures and training sets. Moreover, this view yields a simple and fast method of generating adversarial examples. Using this approach to provide examples for adversarial training, we reduce the test set error of a maxout network on the MNIST dataset.


discusses Adversarial machine learning

Adversarial machine learning is a machine learning technique that attempts to fool models by supplyin...

Edit details Edit relations Attach new author Attach new topic Attach new resource
0.0 /10
useless alright awesome
from 0 reviews
Write comment Rate resource Tip: Rating is anonymous unless you also write a comment.
Resource level 0.0 /10
beginner intermediate advanced
Resource clarity 0.0 /10
hardly clear sometimes unclear perfectly clear
Reviewer's background 0.0 /10
none basics intermediate advanced expert
Comments 0
Currently, there aren't any comments.