Research map / Graph learning / Graph algorithms and clustering
The DFS Fused Lasso: Linear-Time Denoising over General Graphs
Oscar Hernan Madrid Padilla, James Sharpnack, James G. Scott, 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 |
|---|---|---|---|
| use DFS fused lasso instead of Laplacian smoothing | road network denoising · California road network | time for 20 tuning values (s): 2748.26 -> 82.67 | Improved |
| use DFS fused lasso instead of Laplacian smoothing | road network denoising · Pennsylvania road network | time for 20 tuning values (s): 1891.97 -> 44.02 | Improved |
| use DFS fused lasso instead of Laplacian smoothing | road network denoising · Texas road network | time for 20 tuning values (s): 1487.36 -> 54.49 | Improved |
| use DFS fused lasso instead of Laplacian smoothing at high SNR | road network denoising · road networks, piecewise constant signal | optimized MSE | Improved |
| use DFS fused lasso instead of Laplacian smoothing at low SNR | road network denoising · road networks, low total variation | optimized MSE | No clear effect |
| use full 2d fused lasso instead of DFS fused lasso | 2d grid denoising · 2d grid graphs | slope of log MSE vs n: -0.39 -> -0.58 | Improved |
| average 5 random DFS orderings instead of 1 (ablation) | DFS fused lasso · 2d grid graphs | optimized MSE | Improved |
| average 5 random DFS orderings instead of 1 (ablation) | DFS fused lasso · random tree graphs | optimized MSE | Improved |
| use DFS fused lasso instead of wavelet smoothing | tree denoising · random tree graphs | optimized MSE | Improved |
Closest papers by meaning
- Variance estimation in graphs with the fused lasso (JMLR 2024 · similarity 0.92)
- Total Variation Classes Beyond 1d: Minimax Rates, and the Limitations of Linear Smoothers (NeurIPS 2016 · similarity 0.87)
- Equivalence of Graphical Lasso and Thresholding for Sparse Graphs (JMLR 2016 · similarity 0.87)
- Rethinking Graph Transformers as Graph Signal Denoisers: The Role of Block-Diagonal Priors (ICML 2026 · similarity 0.87)
- Is Noise Conditioning Necessary? A Unified Theory of Unconditional Graph Diffusion Models (NeurIPS 2025 · similarity 0.86)
- Split LBI: An Iterative Regularization Path with Structural Sparsity (NeurIPS 2016 · 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.
