Research map / Graph learning / GNN architectures
Fiedler Regularization: Learning Neural Networks with Graph Sparsity
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Replace L1, weight decay, or Dropout with Fiedler-value regularization | 5-layer fully connected NN · MNIST | testing accuracy: 90.25$\pm$ 0.35, 94.98$\pm$ 0.07, 94.5$\pm$ 0.18 -> 96.1$\pm$ 0.12 | Helped |
| Replace L1, weight decay, or Dropout with Fiedler-value regularization | 5-layer fully connected NN · CIFAR10 | testing accuracy: 28.88$\pm$ 1.11, 50.55$\pm$ 0.39, 44.63$\pm$ 0.39 -> 52.26$\pm$ 0.27 | Helped |
| Replace L1 with Fiedler-value regularization | 5-layer fully connected NN · TCGA Pan-Cancer | testing accuracy: 94.53$\pm$ 12.46 -> 90.55$\pm$ 20.73 | No clear effect |
| Replace weight decay or Dropout with Fiedler-value regularization | 5-layer fully connected NN · TCGA Pan-Cancer | testing accuracy: 60.2$\pm$ 20.94, 23.88$\pm$ 7.36 -> 90.55$\pm$ 20.73 | Helped |
Closest papers by meaning
- Rethinking Graph Regularization for Graph Neural Networks (AAAI 2021 · similarity 0.87)
- Learning sparse neural networks via sensitivity-driven regularization (NeurIPS 2018 · similarity 0.87)
- Cauchy-Schwarz Regularizers (ICLR 2025 · similarity 0.86)
- Global Minimizers of $\ell^p$-Regularized Objectives Yield the Sparsest ReLU Neural Networks (NeurIPS 2025 · similarity 0.85)
- Deep Weight Factorization: Sparse Learning Through the Lens of Artificial Symmetries (ICLR 2025 · similarity 0.85)
- Elliptic Loss Regularization (ICLR 2025 · similarity 0.85)
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.