AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Trade-off Between Efficiency and Consistency for Removal-based Explanations

Yifan Zhang, Haowei He, Zhiquan Tan, Yang Yuan

NeurIPS 2023 · Learning theory and algorithms › Prediction and decision losses · approach: Provable post-hoc explanations · 1 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 Harmonica 3rd instead of the strongest listed baselineRemoval-based explanations · IMDbC^3 truthful gap: 1.390 -> 0.174Improved
Use Harmonica 3rd instead of the strongest listed baselineRemoval-based explanations · ImageNetC^3 truthful gap: 0.430 -> 0.246Improved
Use Harmonica-anchor with k=9 instead of Integrated GradientsImageNet explanations · ImageNetL2 interpretation error: 31.8 times lowerImproved
Increase the number of anchors (ablation)Harmonica-anchor · IMDb, ImageNet, and SST-2L2 interpretation errorImproved
Fit separate local models across neighborhood subspaces (ablation)Harmonica-anchor · IMDb, ImageNet, and SST-2L2 interpretation errorImproved
Restrict Harmonica-local to a local neighborhood (ablation)Harmonica-local · Local neighborhood and outside neighborhoodInterpretation errorMixed
Increase the anchor-consistency penalty (ablation)Harmonica-anchor-constrained · Removal-based explanationsInterpretation errorHurt
Use Harmonica 3rd rather than Harmonica 2nd (ablation)Harmonica · IMDbC^3 truthful gap: 0.540 -> 0.174Improved
Use Harmonica 3rd rather than Harmonica 2nd (ablation)Harmonica · ImageNetC^3 truthful gap: 0.660 -> 0.246Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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