Distilling Influences to Mitigate Prediction Churn in Graph Neural Networks.

Andreas Roth, Thomas Liebig

Browse the full ACML paper archive.