AI Paper Atlas

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

ProtoShotXAI: Using Prototypical Few-Shot Architecture for Explainable AI

Samuel Hess, Gregory Ditzler

JMLR 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Post-hoc explanation analysis · 0 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
average the support set into a class prototype instead of taking the best-matching single exemplar (ablation)ProtoShotXAI · MNIST revolving sixagreement with model predicted classImproved
multiply features by classification weights of the class (ablation)ProtoShotXAI · feature attribution mapsnegative attributionsImproved
use Euclidean distance instead of cosine similarity (ablation)ProtoShotXAI · revolving sixinterpretability and consistency with model responseHurt
compute distance at the final classification layer instead of the class-weighted feature layer (ablation)ProtoShotXAI · feature attribution mapsattribution interpretabilityHurt
add few-shot training of adaptive weights between prototype and query features (ablation)ProtoShotXAI · feature attribution mapsattribution qualityNo clear effect
increase support set size from 1 to about 60 samples (ablation)ProtoShotXAI · ImageNet attribution (elephant and zebra)attribution stability: 1 to 220 samples varied; negligible difference by 60; 100 usedImproved
use zero reference pixel instead of average value reference (ablation)ProtoShotXAI · ImageNet tank/bulldog imageattribution magnitude in backgroundMixed
increase perturbation pad size from 1 to 50 (ablation)ProtoShotXAI · ImageNet attributionattribution coverage vs detail: pad 1 to 50; pad of 15 usedMixed
use ResNet50 with Stylized-ImageNet weights instead of ImageNet weightsResNet50 · tank/bulldog imageevidence of shape bias in attributions: 74.9% -> 43.4% accuracyNo clear effect
threshold ProtoShotXAI in-class score to detect adversarial samplesProtoShotXAI · Adversarial MNIST (FGSM, 1,000 samples)detection ROCImproved

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