TY - GEN
T1 - ADMP-GNN
T2 - 34th ACM International Conference on Information and Knowledge Management, CIKM 2025
AU - Abbahaddou, Yassine
AU - Malliaros, Fragkiskos D.
AU - Lutzeyer, Johannes F.
AU - Vazirgiannis, Michalis
N1 - Publisher Copyright:
© 2025 ACM.
PY - 2025/11/10
Y1 - 2025/11/10
N2 - Graph Neural Networks (GNNs) have proven to be highly effective in various graph learning tasks. A key characteristic of GNNs is their use of a fixed number of message-passing steps for all nodes in the graph, regardless of each node's diverse computational needs and characteristics. Through empirical real-world data analysis, we demonstrate that the optimal number of message-passing layers varies for nodes with different characteristics. This finding is further supported by experiments conducted on synthetic datasets. To address this, we propose Adaptive Depth Message Passing GNN (ADMP-GNN), a novel framework that dynamically adjusts the number of message passing layers for each node, resulting in improved performance. This approach applies to any model that follows the message passing scheme. We evaluate ADMP-GNN on the node classification task and observe performance improvements over baseline GNN models. Our code is publicly available at: https://github.com/abbahaddou/ADMP-GNN
AB - Graph Neural Networks (GNNs) have proven to be highly effective in various graph learning tasks. A key characteristic of GNNs is their use of a fixed number of message-passing steps for all nodes in the graph, regardless of each node's diverse computational needs and characteristics. Through empirical real-world data analysis, we demonstrate that the optimal number of message-passing layers varies for nodes with different characteristics. This finding is further supported by experiments conducted on synthetic datasets. To address this, we propose Adaptive Depth Message Passing GNN (ADMP-GNN), a novel framework that dynamically adjusts the number of message passing layers for each node, resulting in improved performance. This approach applies to any model that follows the message passing scheme. We evaluate ADMP-GNN on the node classification task and observe performance improvements over baseline GNN models. Our code is publicly available at: https://github.com/abbahaddou/ADMP-GNN
KW - adaptive neural networks
KW - graph neural networks
UR - https://www.scopus.com/pages/publications/105023140487
U2 - 10.1145/3746252.3761009
DO - 10.1145/3746252.3761009
M3 - Conference contribution
AN - SCOPUS:105023140487
T3 - CIKM 2025 - Proceedings of the 34th ACM International Conference on Information and Knowledge Management
SP - 4
EP - 13
BT - CIKM 2025 - Proceedings of the 34th ACM International Conference on Information and Knowledge Management
PB - Association for Computing Machinery, Inc
Y2 - 10 November 2025 through 14 November 2025
ER -