AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Interpretable Basis Decomposition for Visual Explanation

Bolei Zhou, Yiyou Sun, David Bau, Antonio Torralba

ECCV 2018 · Label-efficient and robust learning › Recognition and long-tail · approach: Self-explaining and visual explanation networks · 276 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use deeper backbone for concept classifier features (AlexNet to ResNet50)Concept classifiers · Broden validationmAP: 0.625 -> 0.804Improved
explain with IBD instead of CAM, Grad-CAM and Network DissectionInterpretable Basis Decomposition · AMT pairwise comparison, four Places365 networkshuman preference (5-point Likert)Improved
use IBD explanations to compare ResNet50 against shallower networks on a single imageInterpretable Basis Decomposition · AMT model-trust comparisonhuman preferenceImproved
IBD compared with Grad-CAM for judging model trustInterpretable Basis Decomposition · AMT model-trust comparisonhuman preferenceNo clear effect
use CAM single heatmap instead of IBD for judging trustCAM · AMT model-trust comparisonhuman preferenceHurt
longer multi-heatmap explanations versus single heatmapInterpretable Basis Decomposition · AMT comparison vs CAM and Grad-CAMhuman preferenceMixed
compare two identical ResNet50 (one mirrored) with IBD or CAMInterpretable Basis Decomposition · AMT model-trust comparisonhuman preferenceNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy