Research map / Learning theory and algorithms / Prediction and decision losses
A Distributed Method for Fitting Laplacian Regularized Stratified Models
Jonathan Tuck, Shane Barratt, Stephen Boyd
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 Laplacian regularization across strata (stratified vs separate model) (ablation) | logistic regression · Mesothelioma classification | Test error: 0.45 -> 0.27 | Improved |
| add stratification (stratified vs common model) (ablation) | logistic regression · Mesothelioma classification | Test ANLL: 0.73 -> 0.76 | No clear effect |
| add stratification (stratified vs common model) (ablation) | logistic regression · Mesothelioma classification | Test error: 0.30 -> 0.27 | Improved |
| add Laplacian regularization across geographic strata (stratified vs separate) (ablation) | ordinary regression · House price prediction (King County) | Test RMSE: 1.514 -> 0.181 | Improved |
| add geographic stratification (stratified vs common model) (ablation) | ordinary regression · House price prediction (King County) | Test RMSE: 0.316 -> 0.181 | Improved |
| use stratified model instead of random forest | ordinary regression · House price prediction (King County) | Test RMSE: 0.184 -> 0.181 | Improved |
| add Laplacian regularization across state-year strata (stratified vs separate) (ablation) | Bernoulli model · Senate elections | Test ANLL: 1.00 -> 0.61 | Improved |
| add stratification (stratified vs common model) (ablation) | Bernoulli model · Senate elections | Test ANLL: 0.70 -> 0.61 | Improved |
| add Laplacian regularization across location-time strata (stratified vs separate) (ablation) | Poisson model · Chicago crime prediction | Test ANLL: 0.740 -> 0.234 | Improved |
| add stratification (stratified vs common model) (ablation) | Poisson model · Chicago crime prediction | Test ANLL: 0.278 -> 0.234 | Improved |
| add Laplacian regularization (stratified vs separate) on training fit (ablation) | Poisson model · Chicago crime prediction | Train ANLL: 0.068 -> 0.221 | Hurt |
Closest papers by meaning
- Feature Grouping as a Stochastic Regularizer for High-Dimensional Structured Data (ICML 2019 · similarity 0.85)
- Overcoming the curse of dimensionality with Laplacian regularization in semi-supervised learning (NeurIPS 2021 · similarity 0.84)
- Estimation of Graphical Models through Structured Norm Minimization (JMLR 2018 · similarity 0.83)
- Learning Local Neighborhoods of Non-Gaussian Graphical Models (AAAI 2025 · similarity 0.82)
- Balancing Act: Distribution-Guided Debiasing in Diffusion Models (CVPR 2024 · similarity 0.82)
- Nonlinear Laplacians: Tunable principal component analysis under directional prior information (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.
