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 convolutional networks have been quite successful at various imageclassification tasks. The current methods to explain the predictions of apre-trained model rely on gradient information, often resulting in saliencymaps that focus on the foreground object as a whole. However, humans typicallyreason by dissecting an image and pointing out the presence of smallerconcepts. The final output is often an aggregation of the presence or absenceof these smaller concepts. In this work, we propose MACE: a Model AgnosticConcept Extractor, which can explain the working of a convolutional networkthrough smaller concepts. The MACE framework dissects the feature mapsgenerated by a convolution network for an image to extract concept basedprototypical explanations. Further, it estimates the relevance of the extractedconcepts to the pre-trained model's predictions, a critical aspect required forexplaining the individual class predictions, missing in existing approaches. Wevalidate our framework using VGG16 and ResNet50 CNN architectures, and ondatasets like Animals With Attributes 2 (AWA2) and Places365. Our experimentsdemonstrate that the concepts extracted by the MACE framework increase thehuman interpretability of the explanations, and are faithful to the underlyingpre-trained black-box model.

MACE: Model Agnostic Concept Extractor for Explaining Image Classification Networks