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

Explaining the decisions of models is becoming pervasive in the imageprocessing domain, whether it is by using post-hoc methods or by creatinginherently interpretable models. While the widespread use of surrogateexplainers is a welcome addition to inspect and understand black-box models,assessing the robustness and reliability of the explanations is key for theirsuccess. Additionally, whilst existing work in the explainability fieldproposes various strategies to address this problem, the challenges of workingwith data in the wild is often overlooked. For instance, in imageclassification, distortions to images can not only affect the predictionsassigned by the model, but also the explanation. Given a clean and a distortedversion of an image, even if the prediction probabilities are similar, theexplanation may still be different. In this paper we propose a methodology toevaluate the effect of distortions in explanations by embedding perceptualdistances that tailor the neighbourhoods used to training surrogate explainers.We also show that by operating in this way, we can make the explanations morerobust to distortions. We generate explanations for images in the Imagenet-Cdataset and demonstrate how using a perceptual distances in the surrogateexplainer creates more coherent explanations for the distorted and referenceimages.

Explainers in the Wild: Making Surrogate Explainers Robust to Distortions through Perception