AI Paper Atlas

Research map / Graph learning / GNN architectures

Robust Counterfactual Explanations on Graph Neural Networks

Mohit Bajaj, Lingyang Chu, Zi Yu Xue, Jian Pei, Lanjun Wang, Peter Cho-Ho Lam, Yong Zhang

NeurIPS 2021 · Graph learning › GNN architectures · approach: GNN explainability · 33 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 decision-boundary (LDB) loss instead of direct confidence loss (ablation)RCExplainer · Mutagenicity, NCI1, BA-2motifsfidelityImproved
add decision-boundary (LDB) loss instead of direct confidence loss (ablation)RCExplainer · BA-2motifs, NCI1noise robustness AUCImproved
increase sampled LDBs from 10 to 50 (ablation)RCExplainer · BA-CommunityfidelityImproved
lower lambda to 0.1 (more weight on counterfactual loss) instead of 0.9 (ablation)RCExplainer · BA-2motifsfidelityImproved
replace per-graph explanation optimisation with decision-region-based explanation networkRCExplainer vs GNNExplainer · BA-Shapesexplanation AUC: 0.925 -> 0.998Improved
replace PGM-Explainer with RCExplainerRCExplainer · BA-Communitynode accuracy: 0.926 -> 0.916Hurt
use RCExplainer instead of PGExplainerRCExplainer · Mutagenicitynoise robustness AUCNo clear effect
use trained explanation network without retrainingRCExplainer · Mutagenicityexplanation time (s): 77.8s -> 0.01sImproved

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