Minesweeper
Description
Minesweeper
Dataset Overview
Lifting Methods
Structural-based Liftings
- Cellular: Cycle-based lifting
- Simplicial: Clique complex lifting
- Hypergraph: k-hop lifting
Feature Lifting
- Projected sum
Model Performance
| Model | Accuracy (%) | Std Dev (±) |
|---|---|---|
| GCN | 87.52 | 0.42 |
| GIN | 87.82 | 0.34 |
| GAT | 89.64 | 0.43 |
| AST | 81.14 | 0.05 |
| EDGNN | 84.52 | 0.05 |
| UniGNN2 | 78.02 | 0.00 |
| CWN | 88.62 | 0.04 |
| CCCN | 89.42 | 0.00 |
| SCCNN | 89.0 | 0.00 |
| SCN | 90.32 | 0.11 |
Key Insights
- SCN achieves the best performance with 90.32% accuracy
- GAT, CCCN, and SCCNN also perform strongly, all above 89%
- Most models perform above 84%, but UniGNN2 and AST are notably lower
- GIN and GCN show consistent results with low standard deviation
Reproducibility
python -m topobench model=simplicial/scn dataset=graph/minesweeper optimizer.parameters.lr=0.01 model.feature_encoder.out_channels=64 model.backbone.n_layers=3 model.readout.readout_name=NoReadOut transforms.graph2simplicial_lifting.signed=True model.feature_encoder.proj_dropout=0.5 dataset.dataloader_params.batch_size=1 dataset.split_params.data_seed=0,3,5,7,9 trainer.max_epochs=1000 trainer.min_epochs=50 trainer.check_val_every_n_epoch=1 callbacks.early_stopping.patience=50 logger.wandb.project=TopoBench --multirun