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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Reliable deployment of machine learning models such as neural networkscontinues to be challenging due to several limitations. Some of the mainshortcomings are the lack of interpretability and the lack of robustnessagainst adversarial examples or out-of-distribution inputs. In this exploratoryreview, we explore the possibilities and limits of adversarial attacks forexplainable machine learning models. First, we extend the notion of adversarialexamples to fit in explainable machine learning scenarios, in which the inputs,the output classifications and the explanations of the model's decisions areassessed by humans. Next, we propose a comprehensive framework to study whether(and how) adversarial examples can be generated for explainable models underhuman assessment, introducing and illustrating novel attack paradigms. Inparticular, our framework considers a wide range of relevant yet often ignoredfactors such as the type of problem, the user expertise or the objective of theexplanations, in order to identify the attack strategies that should be adoptedin each scenario to successfully deceive the model (and the human). Theintention of these contributions is to serve as a basis for a more rigorous andrealistic study of adversarial examples in the field of explainable machinelearning.

When and How to Fool Explainable Models (and Humans) with Adversarial Examples