English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep learning has greatly improved visual recognition in recent years.However, recent research has shown that there exist many adversarial examplesthat can negatively impact the performance of such an architecture. This paperfocuses on detecting those adversarial examples by analyzing whether they comefrom the same distribution as the normal examples. Instead of directly traininga deep neural network to detect adversarials, a much simpler approach wasproposed based on statistics on outputs from convolutional layers. A cascadeclassifier was designed to efficiently detect adversarials. Furthermore,trained from one particular adversarial generating mechanism, the resultingclassifier can successfully detect adversarials from a completely differentmechanism as well. The resulting classifier is non-subdifferentiable, hencecreates a difficulty for adversaries to attack by using the gradient of theclassifier. After detecting adversarial examples, we show that many of them canbe recovered by simply performing a small average filter on the image. Thosefindings should lead to more insights about the classification mechanisms indeep convolutional neural networks.

Adversarial Examples Detection in Deep Networks with Convolutional Filter Statistics