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

Image classifiers are an important component of today's software, fromconsumer and business applications to safety-critical domains. The advent ofDeep Neural Networks (DNNs) is the key catalyst behind such wide-spreadsuccess. However, wide adoption comes with serious concerns about therobustness of software systems dependent on DNNs for image classification, asseveral severe erroneous behaviors have been reported under sensitive andcritical circumstances. We argue that developers need to rigorously test theirsoftware's image classifiers and delay deployment until acceptable. We presentan approach to testing image classifier robustness based on class propertyviolations.  We found that many of the reported erroneous cases in popular DNN imageclassifiers occur because the trained models confuse one class with another orshow biases towards some classes over others. These bugs usually violate someclass properties of one or more of those classes. Most DNN testing techniquesfocus on per-image violations, so fail to detect class-level confusions orbiases.  We developed a testing technique to automatically detect class-basedconfusion and bias errors in DNN-driven image classification software. Weevaluated our implementation, DeepInspect, on several popular image classifierswith precision up to 100% (avg.~72.6%) for confusion errors, and up to 84.3%(avg.~66.8%) for bias errors. DeepInspect found hundreds of classificationmistakes in widely-used models, many exposing errors indicating confusion orbias.

Testing DNN Image Classifiers for Confusion & Bias Errors