aiqbench analysis
This research introduces MEGA-GNN, a novel framework for processing edge-attributed multigraphs that resolves limitations in handling repeated interactions while maintaining permutation equivariance. By implementing neighbor-aware aggregation, the model captures per-neighbor statistics essential for complex systems like social and financial networks without increasing asymptotic complexity. This advancement represents a significant step forward in graph neural network architectures for modeling intricate real-world data structures.
MEGA: Message Passing Neural Networks for Multigraphs with EdGe Attributes
arXiv:2412.00241v3 Announce Type: replace Abstract: Edge-attributed multigraphs, in which multiple edges with distinct attributes connect the same pair of nodes, arise naturally in many real-world systems. In these graphs, effective learning requires preserving information from repeated interactions while distinguishing contributions from different neighbors. Existing neural network solutions for edge-attributed multigraphs remain limited: some lose information from repeated interactions, while others break permutation equivariance. To address this, we introduce \emph{neighbor-aware aggregation}, an operator that first combines multi-edge features for each neighbor and then aggregates across neighbors. This operator captures per-neighbor statistics that standard single-stage aggregation cannot represent. Building on this operator, we present MEGA-GNN, a model-agnostic message-passing framework for edge-attributed multigraphs. We show that MEGA-GNN is permutation equivariant and has the same asymptotic complexity as standard GNNs with edge updates. We evaluate our approach on datasets from social networks and financial transaction networks. Neighbor-aware aggregation consistently improves GNN performance and matches or surpasses state-of-the-art methods.
Abstract:Edge-attributed multigraphs, in which multiple edges with distinct attributes connect the same pair of nodes, arise naturally in many real-world systems. In these graphs, effective learning requires preserving information from repeated interactions while distinguishing contributions from different neighbors. Existing neural network solutions for edge-attributed multigraphs remain limited: some lose information from repeated interactions, while others break permutation equivariance. To address this, we introduce \emph{neighbor-aware aggregation}, an operator that first combines multi-edge features for each neighbor and then aggregates across neighbors. This operator captures per-neighbor statistics that standard single-stage aggregation cannot represent. Building on this operator, we present MEGA-GNN, a model-agnostic message-passing framework for edge-attributed multigraphs. We show that MEGA-GNN is permutation equivariant and has the same asymptotic complexity as standard GNNs with edge updates. We evaluate our approach on datasets from social networks and financial transaction networks. Neighbor-aware aggregation consistently improves GNN performance and matches or surpasses state-of-the-art methods.
Machine Learning (cs.LG)
arXiv:2412.00241 [cs.LG]
(or arXiv:2412.00241v3 [cs.LG] for this version)
From: H. Çağrı Bilgi [view email] [v1] Fri, 29 Nov 2024 20:15:18 UTC (495 KB) [v2] Tue, 10 Dec 2024 08:27:38 UTC (509 KB) [v3] Mon, 31 Aug 2026 13:59:21 UTC (3,077 KB)
Full text under the publisher's open licence · 193 words
Related coverage
Privacy-Preserving GNN Inference on Financial Transaction Graphs via Homomorphic Encryption
arXiv · Oct 14, 2026
Benchmarking Sheaf Neural Networks for Inductive Tasks
arXiv · Aug 4, 2026
DuoGNN: Topology-aware Graph Neural Network with Homophily and Heterophily Interaction-Decoupling
arXiv · Aug 25, 2026
Subgraph Filtering for Fair Graph Neural Networks
arXiv · Aug 28, 2026