Research map / Probabilistic and scientific modeling / Molecules and proteins
Modelling Interactions in High-dimensional Data with Backtracking
Rajen D. Shah
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 Backtracking instead of Lasso on main effects only | Lasso · Simulation scenario 3 (SNR=2) | L2-sq: 6.95 -> 1.21 | Improved |
| use Backtracking instead of Lasso on main effects only | Lasso · Simulation scenario 5 (SNR=3) | L2-sq: 12.84 -> 1.17 | Improved |
| use Backtracking instead of iterated Lasso | Lasso · Simulation scenario 4 (SNR=2) | L2-sq: 3.25 -> 2.72 | Improved |
| use Backtracking instead of iterated Lasso | Lasso · Simulation scenario 5 (SNR=3) | FP Inter: 3.05 -> 0.55 | Improved |
| use Backtracking instead of marginal screening of interactions | Lasso · Simulation scenario 4 (SNR=3) | L2-sq: 15.19 -> 0.41 | Improved |
| use Backtracking instead of iterated Lasso at SNR 3 in scenario 3 | Lasso · Simulation scenario 3 (SNR=3) | L2-sq: 0.27 -> 0.27 | No clear effect |
| add marginal screening of 2p interactions before the Lasso | Lasso · Simulation scenario 1 | prediction error | Hurt |
| use Backtracking instead of main effects Lasso | Lasso · Communities and Crime | test error: 0.414 -> 0.365 | Improved |
| use Backtracking instead of iterated Lasso | Lasso · Communities and Crime | test error: 0.384 -> 0.365 | Improved |
| use Backtracking instead of Random Forest | Lasso · Communities and Crime | test error: 0.356 -> 0.365 | Hurt |
| use Backtracking instead of main effects multinomial Lasso | multinomial Lasso · ISOLET E-set | misclassification error: 0.0641 -> 0.0563 | Improved |
| use Backtracking instead of Random Forest | multinomial Lasso · ISOLET E-set | misclassification error: 0.0837 -> 0.0563 | Improved |
Closest papers by meaning
- The SKIM-FA Kernel: High-Dimensional Variable Selection and Nonlinear Interaction Discovery in Linear Time (JMLR 2023 · similarity 0.88)
- High-Dimensional Interactions Detection with Sparse Principal Hessian Matrix (JMLR 2020 · similarity 0.87)
- The Kernel Interaction Trick: Fast Bayesian Discovery of Pairwise Interactions in High Dimensions (ICML 2019 · similarity 0.87)
- The Complete Lasso Tradeoff Diagram (NeurIPS 2020 · similarity 0.86)
- The xyz algorithm for fast interaction search in high-dimensional data (JMLR 2018 · similarity 0.86)
- Low-rank Interaction with Sparse Additive Effects Model for Large Data Frames (NeurIPS 2018 · 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.
