Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Dependent relevance determination for smooth and structured sparse regression
Anqi Wu, Oluwasanmi Koyejo, Jonathan Pillow
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 |
|---|---|---|---|
| add smoothness (smooth-DRD) to DRD prior | DRD · Simulated weights, R2 of w | R2 | Improved |
| use Laplace approximation instead of full MCMC for smooth-DRD | smooth-DRD · Simulated weights, R2 of w | R2 | No clear effect |
| add smoothness/structured sparsity (DRD) versus lasso, ARD, TV-L1, GraphNet, SCAD | lasso/ARD · Simulated weights, R2 of w | R2 | Improved |
| use DRD with smooth prior instead of ASD | ASD · Simulated, support recovery | AUC | Improved |
| use MCMC instead of Laplace inference for DRD | DRD-Laplace · Simulated, support recovery | AUC | Hurt |
| cluster non-zero coefficients and use DRD instead of lasso | lasso · Phase transition, p=500 | measurement rate for perfect recovery | Improved |
| add smoothness (sDRD) to DRD for smooth sparse signals | DRD · Phase transition, p=500 | measurement rate for perfect recovery | Improved |
| use DRD instead of lasso, GraphNet, TV-L1 | lasso/GraphNet/TV-L1 · fMRI gambling task, inter-subject | test R2 | Improved |
| use DRD/smooth-DRD instead of lasso, GraphNet, TV-L1 | lasso/GraphNet/TV-L1 · OASIS age prediction | MAE | Improved |
| add smoothness to DRD | DRD · OASIS age prediction | MAE | No clear effect |
| use smooth-DRD instead of DRD for inference cost | DRD · Running time vs p | running time | Improved |
Closest papers by meaning
- Joint Model and Data Sparsification via the Marginal Likelihood (ICML 2026 · similarity 0.89)
- Fast Stagewise Sparse Factor Regression (JMLR 2022 · similarity 0.87)
- Boosted Sparse and Low-Rank Tensor Regression (NeurIPS 2018 · similarity 0.87)
- Robust Gaussian Processes via Relevance Pursuit (NeurIPS 2024 · similarity 0.86)
- Training Bayesian Neural Networks with Sparse Subspace Variational Inference (ICLR 2024 · similarity 0.85)
- Efficient Network Automatic Relevance Determination (ICML 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.
