Research map / Graph learning / GNN architectures
Directed Semi-Simplicial Learning with Applications to Brain Activity Decoding
Manuel Lecha, Andrea Cavallo, Francesca Dominici, Ran Levi, Alessio Del Bue, Elvin Isufi, Pietro Morerio, Claudio Battiloro
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use SSN instead of best baseline | DirGNN-256 · (4,125um) fixed | accuracy (%): 50.89 -> 75.13 | Improved |
| use SSN instead of best baseline | DirGNN-256 · (4,325um) fixed | accuracy (%): 60.02 -> 87.16 | Improved |
| use SSN instead of MPSNN-64 | MPSNN-64 · (8,175um) fixed | accuracy (%): 64.02 -> 78.32 | Improved |
| use SSN instead of MPSNN-64 | MPSNN-64 · (4,325um) M=1 | accuracy (%): 29.48 -> 46.73 | Improved |
| use SSN instead of DirGNN-256 | DirGNN-256 · (4,325um) M=3 | accuracy (%): 35.21 -> 61.35 | Improved |
| use SSN instead of MPSNN-64 | MPSNN-64 · (4,325um) M=5 | accuracy (%): 42.23 -> 64.72 | Improved |
| use routing (top-k relations) R-SSN instead of full SSN | SSN · (4,325um) M=1 | accuracy (%): 46.73 -> 28.68 | Hurt |
| use R-SSN instead of full SSN | SSN · (4,325um) fixed | accuracy (%): 87.16 -> 79.64 | Hurt |
| scale GNN hidden size to 256 | GNN · (4,325um) fixed | accuracy (%): 21.56 -> 23.02 | No clear effect |
| scale DirGNN hidden size to 256 | DirGNN · (4,325um) fixed | accuracy (%): 48.11 -> 60.02 | Improved |
| scale MPSNN hidden size to 64 | MPSNN · (4,325um) fixed | accuracy (%): 42.65 -> 54.24 | Improved |
| use hand-crafted topological features with SVM instead of GNN | GNN-256 · (4,125um) fixed | accuracy (%): 24.70 -> 42.14 | Improved |
Closest papers by meaning
- TopoTune: A Framework for Generalized Combinatorial Complex Neural Networks (ICML 2025 · similarity 0.89)
- Topological Blindspots: Understanding and Extending Topological Deep Learning Through the Lens of Expressivity (ICLR 2025 · similarity 0.88)
- E(n) Equivariant Topological Neural Networks (ICLR 2025 · similarity 0.87)
- Self-Supervised Discovery of Neural Circuits in Spatially Patterned Neural Responses with Graph Neural Networks (NeurIPS 2025 · similarity 0.86)
- Topological Neural Networks go Persistent, Equivariant, and Continuous (ICML 2024 · similarity 0.86)
- From Latent Graph to Latent Topology Inference: Differentiable Cell Complex Module (ICLR 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
