MEGA-GNN adds a parallel-edge aggregation stage inside each message passing layer, giving multigraph GNNs permutation equivariance and, when edge features are totally ordered, universality, with up to 13 percentage point F1 gains on money laundering detection.
Efficient subgraph GNN s by learning effective selection policies
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Multigraph Message Passing with Bi-Directional Multi-Edge Aggregations
MEGA-GNN adds a parallel-edge aggregation stage inside each message passing layer, giving multigraph GNNs permutation equivariance and, when edge features are totally ordered, universality, with up to 13 percentage point F1 gains on money laundering detection.