Research map / Graph learning / Graph algorithms and clustering
Trend Filtering on Graphs
Yu-Xiang Wang, James Sharpnack, Alexander J. Smola, Ryan J. Tibshirani
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 |
|---|---|---|---|
| replace Laplacian regularizer with graph trend filtering k=2 | MAD label imputation · UCI wine | misclassification rate: 0.060 -> 0.052 | Improved |
| replace Laplacian regularizer with graph trend filtering k=2 | MAD label imputation · UCI breast | misclassification rate: 0.064 -> 0.059 | Improved |
| replace Laplacian regularizer with graph trend filtering k=2 | MAD label imputation · UCI yeast | misclassification rate: 0.566 -> 0.552 | Improved |
| replace Laplacian regularizer with graph trend filtering k=2 | MAD label imputation · UCI iris | misclassification rate: 0.085 -> 0.084 | No clear effect |
| replace Laplacian regularizer with graph trend filtering k=0 | MAD label imputation · UCI breast | misclassification rate: 0.064 -> 0.500 | Hurt |
| replace Laplacian regularizer with graph trend filtering k=0 | MAD label imputation · UCI heart | misclassification rate: 0.208 -> 0.472 | Hurt |
| replace Laplacian regularizer with graph trend filtering k=0 | MAD label imputation · UCI ads | misclassification rate: 0.306 -> 0.212 | Improved |
| replace Laplacian regularizer with graph trend filtering k=1 | MAD label imputation · UCI car | misclassification rate: 0.316 -> 0.293 | Improved |
| use graph trend filtering (k=2) instead of Laplacian smoothing at low degrees of freedom | denoising · Allegheny County example | mean squared error: lower MSE at low df; matches Laplacian when overfitting | Improved |
| use spanning tree wavelet design instead of other wavelet designs (ablation) | wavelet smoothing · Allegheny County and Facebook examples | mean squared error | Improved |
Closest papers by meaning
- Higher-Order Total Variation Classes on Grids: Minimax Theory and Trend Filtering Methods (NeurIPS 2017 · similarity 0.87)
- Enhance Before Fusion: Multi-View Graph Clustering With Graph Trend Filter (TPAMI 2026 · similarity 0.85)
- Variance estimation in graphs with the fused lasso (JMLR 2024 · similarity 0.84)
- Adaptive Estimation of Graphical Models under Total Positivity (ICML 2023 · similarity 0.83)
- LLM-based Online Prediction of Time-varying Graph Signals (Student Abstract) (AAAI 2025 · similarity 0.82)
- HeroFilter: Adaptive Spectral Graph Filter for Varying Heterophilic Relations (NeurIPS 2025 · similarity 0.82)
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.
