Conformal Inductive Graph Neural Networks
This work addresses a key limitation in uncertainty quantification for graph neural networks, enabling reliable predictions in dynamic or inductive scenarios, though it is incremental as it builds on existing conformal prediction frameworks.
The paper tackled the problem of applying conformal prediction to inductive graph neural network settings, where conventional methods fail due to shifts from message passing with new nodes, and achieved a solution that recovers standard coverage guarantees without sacrificing statistical efficiency, with guarantees proven to hold independently of prediction time.
Conformal prediction (CP) transforms any model's output into prediction sets guaranteed to include (cover) the true label. CP requires exchangeability, a relaxation of the i.i.d. assumption, to obtain a valid distribution-free coverage guarantee. This makes it directly applicable to transductive node-classification. However, conventional CP cannot be applied in inductive settings due to the implicit shift in the (calibration) scores caused by message passing with the new nodes. We fix this issue for both cases of node and edge-exchangeable graphs, recovering the standard coverage guarantee without sacrificing statistical efficiency. We further prove that the guarantee holds independently of the prediction time, e.g. upon arrival of a new node/edge or at any subsequent moment.