AI Paper Atlas

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

ICAM: Interpretable Classification via Disentangled Representations and Feature Attribution Mapping

Cher Bass, Mariana da Silva, Carole Sudre, Petru-Daniel Tudosiu, Stephen Smith, Emma Robinson

NeurIPS 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Feature attribution methods · 9 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
add the full ICAM components and losses, including the L2 loss (ablation)ICAM · HCP 2D lesion simulationsNCC (-): 0.385 $\pm$ 0.094 -> \bf 0.435 $\pm$ 0.092Improved
add attribute classification BCE loss and rejection sampling (ablation)ICAM-DRIT · HCP 2D lesion simulationsNCC (+): 0.265 $\pm$ 0.062 -> 0.325 $\pm$ 0.100Improved
add feature attribution map loss (ablation)ICAM-BCE · HCP 2D lesion simulationsNCC (+): 0.325 $\pm$ 0.100 -> \bf 0.333 $\pm$ 0.082Improved
add the full ICAM losses, including L2 loss (ablation)ICAM-FA · HCP 2D lesion simulationsNCC (-): 0.353 $\pm$ 0.104 -> \bf 0.435 $\pm$ 0.092Improved
increase the attribute latent space size from 8 to 80 (ablation)DRIT · HCP 2D lesion simulationsNCC (-): 0.346 $\pm$ 0.080 -> 0.380 $\pm$ 0.081Improved
use ICAM architecture with DRIT losses instead of DRIT z80 (ablation)DRIT z80 · HCP 2D lesion simulationsNCC (-): 0.380 $\pm$ 0.081 -> 0.385 $\pm$ 0.094No clear effect
use full ICAM rather than VA-GANADNI paired MCI-to-AD conversions · ADNI paired MCI-to-AD conversionsNCC (-): $0.653\pm 0.142$ -> \bf 0.683 $\pm$ 0.097Improved
use full ICAM rather than Grad-CAMADNI paired MCI-to-AD conversions · ADNI paired MCI-to-AD conversionsNCC (+): $0.461 \pm 0.086 $ -> \bf 0.652 $\pm$ 0.083Improved
use full ICAM rather than Integrated gradientsADNI paired MCI-to-AD conversions · ADNI paired MCI-to-AD conversionsNCC (-): $0.378 \pm 0.064 $ -> \bf 0.683 $\pm$ 0.097Improved
use full ICAM rather than OcclusionADNI paired MCI-to-AD conversions · ADNI paired MCI-to-AD conversionsNCC (+): $0.354 \pm 0.057 $ -> \bf 0.652 $\pm$ 0.083Improved
use full ICAM rather than ICAM-DRIT (ablation)UK Biobank age classification · UK Biobank age classificationlatent-space class separationImproved
use full ICAM rather than VA-GANUK Biobank ageing translation · UK Biobank ageing translationfeature attribution variabilityImproved

Later papers that compared with this method

1 later accepted paper reports ICAM as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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