New method boosts temporal graph neural networks with motif signatures

Researchers have developed a new method to enhance temporal graph neural networks (TGNNs) by incorporating temporal motif signatures. These signatures capture predictive patterns like repetition and reciprocity within interaction streams, which standard TGNNs often miss. The proposed approach uses a compact feature map that can be integrated into existing TGNN architectures, consistently improving performance across various prediction and classification tasks on both real and synthetic datasets. AI

IMPACT Enhances predictive capabilities of graph neural networks for time-series data.

RANK_REASON The cluster contains an academic paper detailing a new method for temporal graph neural networks. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Dylan Sandfelder
• Temporal Graph Neural Networks
• Temporal Motif Signatures

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →