Research map / Graph learning / GNN architectures
Graph Neural Networks with Convolutional ARMA Filters
Filippo Maria Bianchi, Daniele Grattarola, Lorenzo Livi, Cesare Alippi
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 ARMA layer instead of GCN layer | GNN · PPI | accuracy: 80.8 -> 90.5 | Improved |
| use ARMA layer instead of Chebyshev layer | GNN · PPI | accuracy: 86.4 -> 90.5 | Improved |
| use ARMA layer instead of GCN layer | GNN · Cora | accuracy: 81.5 -> 83.4 | Improved |
| use ARMA layer instead of CayleyNet | GNN · Citeseer | accuracy: 67.1 -> 72.5 | Improved |
| use ARMA layer instead of GCN layer | GNN · Pubmed | accuracy: 79.0 -> 78.9 | No clear effect |
| use ARMA layer instead of GCN layer | GNN · 20news | accuracy: 65.45 -> 70.02 | Improved |
| use ARMA layer instead of GCN layer | GNN · MNIST | accuracy: 98.48 -> 99.20 | Improved |
| use ARMA layer instead of Chebyshev layer | GNN · MNIST | accuracy: 99.14 -> 99.20 | No clear effect |
| use ARMA layer instead of GCN layer | GNN · D&D | accuracy: 74.7 -> 77.6 | Improved |
| use ARMA layer instead of CayleyNet | GNN · MUTAG | accuracy: 87.8 -> 91.5 | Improved |
| use ARMA layer instead of Chebyshev layer | GNN · Enzymes | accuracy: 57.9 -> 60.6 | Improved |
| use ARMA layer instead of CayleyNet | GNN · QM9 U0_atom | MSE: 1.493 -> 0.053 | Improved |
Closest papers by meaning
- Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering (NeurIPS 2016 · similarity 0.90)
- Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs (CVPR 2017 · similarity 0.89)
- Unifying Homophily and Heterophily for Spectral Graph Neural Networks via Triple Filter Ensembles (NeurIPS 2024 · similarity 0.88)
- HoloNets: Spectral Convolutions do extend to Directed Graphs (ICLR 2024 · similarity 0.87)
- SLOG: An Inductive Spectral Graph Neural Network Beyond Polynomial Filter (ICML 2024 · similarity 0.87)
- Graph Neural Networks Are More Than Filters: Revisiting and Benchmarking from A Spectral Perspective (ICLR 2025 · similarity 0.87)
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.
