Research map / Learning theory and algorithms / Prediction and decision losses
Stable Classification
Dimitris Bertsimas, Jack Dunn, Ivan Paskov
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 stable cutting-plane training (SCP) | SVM · 30 UCI datasets | accuracy: 0.773 -> 0.795 | Improved |
| use stable robust counterpart (SRC) | SVM · 30 UCI datasets | accuracy: 0.773 -> 0.807 | Improved |
| use stable robust counterpart (SRC) | SVM · 30 UCI datasets | output stability: 8.493 × 10−3 -> 5.672 × 10−3 | Improved |
| use stable Monte Carlo (SMC) | SVM · 30 UCI datasets | output stability: 8.493 × 10−3 -> 9.567 × 10−3 | Hurt |
| use stable cutting-plane training (SCP) | Logistic Regression · 30 UCI datasets | output stability: 1.004 × 10−2 -> 8.083 × 10−3 | Improved |
| use stable cutting-plane training (SCP) | Logistic Regression · 30 UCI datasets | accuracy: 0.672 -> 0.657 | Hurt |
| use stable robust counterpart (SRC) | Logistic Regression · 30 UCI datasets | accuracy: 0.672 -> 0.668 | No clear effect |
| use stable Monte Carlo (SMC) | Random Forest · 30 UCI datasets | accuracy: 0.847 -> 0.849 | No clear effect |
| use stable cutting-plane training (SCP) | Random Forest · 30 UCI datasets | output stability: 0.016 -> 0.014 | Improved |
| use stable Monte Carlo (SMC) | Random Forest · 30 UCI datasets | hyperparameter stability: 5.924 -> 5.605 | Improved |
| use stable Monte Carlo (SMC) | OCT · 30 UCI datasets | output stability: 0.029 -> 0.026 | Improved |
| use stable training (SMC or SCP) | OCT · 30 UCI datasets | accuracy: 0.828 -> 0.825 | Hurt |
Closest papers by meaning
- On the Stability of Feature Selection Algorithms (JMLR 2018 · similarity 0.89)
- Stable Regression: On the Power of Optimization over Randomization (JMLR 2020 · similarity 0.89)
- Building a stable classifier with the inflated argmax (NeurIPS 2024 · similarity 0.89)
- Learning Stable Classifiers by Transferring Unstable Features (ICML 2022 · similarity 0.89)
- Optimal Classification Trees for Continuous Feature Data Using Dynamic Programming with Branch-and-Bound (AAAI 2025 · similarity 0.86)
- Restoring balance: principled under/oversampling of data for optimal classification (ICML 2024 · similarity 0.84)
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.
