New logic framework extracts compact global rules for graph neural networks

Researchers have developed a new logic-based framework for node classification in Simple Graph Convolution (SGC) networks. This framework utilizes minimal abductive explanations as an intermediate step for extracting global logical rules. By identifying the essential node-feature pairs needed to predict a node's class, the system trains decision trees that yield compact and accurate global rules, outperforming previous methods that relied on potentially redundant explanatory subgraphs. AI

IMPACT This research offers a more efficient method for extracting interpretable rules from graph neural networks, potentially improving their explainability and applicability in complex systems.

RANK_REASON This is a research paper detailing a new framework for node classification in graph neural networks. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• arXiv
• graph neural networks
• LogicXGNN
• Simple Graph Convolution

• paper
• model release

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