Unverified50% confidenceFactExact time
For GNN models with additive message passing and edge coefficients independent of the edge set, structural deletion and message suppression are exactly equivalent interventions.
1
Sources
50%
Confidence
Long-term
Relevance
11/4/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWhen comparing different GNN explanation methods, if each uses a different edge-deletion semantics, the comparison may not be conducted on the same basis.76% similarUnverifiedOn the feeder data, parameter sharing recovers about two-thirds of the error gap between node-specific models and the GNN (relative to the regularized baseline).72% similarUnverifiedGCN's degree-normalization coefficients depend on node degree, and deleting an edge changes the degree and thus the coefficients of the surviving edges, so GCN does not satisfy the equivalence condition of "edge coefficients independent of the edge set."71% similarUnverified图神经网络(GNN)核心操作是消息传递,不同图卷积变体如GCN、GAT、GraphSAGE在不同任务上各有优劣71% similarUnverifiedGNNs address the problems in graph data of a non-fixed number of neighbors and meaningless ordering through the message passing mechanism: in each layer, a node first aggregates the feature information of its neighbors, then combines it with its own representation and updates it through a nonlinear transformation.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1134406API
curl https://kongchang.com/api/v1/knowledge/claims/1134406MCP
get_claim(id=1134406)