Researchers have developed a new dataset of over 131,000 Cayley graphs to serve as benchmarks for studying how finite group properties are reflected in graph observables. This work also contributes new enumerative sequences to the OEIS and identifies empirical regularities, leading to testable conjectures about graph properties. A comparison of classical models, MLPs, and graph neural networks (GNNs) demonstrated that GNNs, particularly GIN and GCN, can effectively predict algebraic group properties directly from graph data. AI
IMPACT Advances the use of GNNs for predicting algebraic properties from graph structures, potentially impacting fields relying on symmetry analysis.
RANK_REASON The cluster contains an academic paper detailing a new dataset and benchmark for studying mathematical properties using machine learning models.
- graph convolutional network
- Graph Information Network
- GNN
- multilayer perceptron
- arXiv
- Cayley Graphs with given Arc-Type
- On-Line Encyclopedia of Integer Sequences
- stat.ML
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
