AslanDing/Robust-Fidelity
a robust metric (robust fidelity) for XGNN (ICLR24)
This project offers a robust method to assess how well Graph Neural Network (GNN) explanations truly reflect why a model made a particular decision. It takes in GNN explanations and outputs quantitative 'fidelity' scores, which indicate the trustworthiness of those explanations, even when facing small changes or noise in the input data. Researchers and practitioners working with explainable AI for graph-based data would use this.
No commits in the last 6 months.
Use this if you need to rigorously evaluate the reliability and robustness of explanations generated by Graph Neural Networks (GNNs).
Not ideal if you are working with non-graph data like images or time series, or if you need to generate GNN explanations rather than evaluate them.
Stars
12
Forks
2
Language
Python
License
—
Last pushed
Jun 03, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/AslanDing/Robust-Fidelity"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
obss/sahi
Framework agnostic sliced/tiled inference + interactive ui + error analysis plots
tensorflow/tcav
Code for the TCAV ML interpretability project
MAIF/shapash
🔅 Shapash: User-friendly Explainability and Interpretability to Develop Reliable and Transparent...
TeamHG-Memex/eli5
A library for debugging/inspecting machine learning classifiers and explaining their predictions
csinva/imodels
Interpretable ML package 🔍 for concise, transparent, and accurate predictive modeling...