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

While deep learning has led to huge progress in complex image classificationtasks like ImageNet, unexpected failure modes, e.g. via spurious features, callinto question how reliably these classifiers work in the wild. Furthermore, forsafety-critical tasks the black-box nature of their decisions is problematic,and explanations or at least methods which make decisions plausible are neededurgently. In this paper, we address these problems by generating images thatoptimize a classifier-derived objective using a framework for guided imagegeneration. We analyze the behavior and decisions of image classifiers byvisual counterfactual explanations (VCEs), detection of systematic mistakes byanalyzing images where classifiers maximally disagree, and visualization ofneurons to verify potential spurious features. In this way, we validateexisting observations, e.g. the shape bias of adversarially robust models, aswell as novel failure modes, e.g. systematic errors of zero-shot CLIPclassifiers, or identify harmful spurious features. Moreover, our VCEsoutperform previous work while being more versatile.

Analyzing and Explaining Image Classifiers via Diffusion Guidance