Machine Learning
7 articles
Regression vs. language-model readouts for survey prediction
An article-reported local comparison of logistic regression and language-model readouts. Scripts, provenance, and matched outputs are not public.
Prediction-powered inference for AI-assisted survey estimation
A method guide with an article-reported HINTS imputation example and a separate regression-adjustment simulation. The local results are not publicly reproduced.
A regression view of steering vectors
An article-reported local experiment comparing steering-vector estimators. The scripts, advance document, and matched outputs are not publicly reproduced.
Imbalanced Classification: An Article-Reported Example
A scikit-learn workflow for imbalanced classification. Dataset counts and model metrics are article-reported; the implementation is not public.
SHAP Classifier Interpretation: An Article-Reported Example
A SHAP workflow for classifier interpretation. AUC, feature counts, runtime, and the referenced implementation are not publicly reproduced.
Can a Classifier Find What Investigators Miss? - Too Early To Say
An evidence-labeled guide to temporal validation, interpretable baselines, and article-reported Medicaid fraud-classifier metrics under reconciliation.
Grocery Store Classifier Results Under Review
The article reports 94% balanced accuracy and 94% and 96% spot-check rates; those metrics and the feature-importance values are not publicly reproduced.