AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Forward-Backward Selection with Early Dropping

Giorgos Borboudakis, Ioannis Tsamardinos

JMLR 2019 · Learning theory and algorithms › Prediction and decision losses · approach: Distributionally robust optimization · 85 citations (OpenAlex)

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 changedWhereMeasuredEffect
Use early dropping with K=0 in forward-backward selectionFBS · Null data, p=200, alpha=0.05number of selected variables (type I errors): 20.8 -> 5.6Improved
Use early dropping with K=1 in forward-backward selectionFBS · Null data, p=200, alpha=0.1number of selected variables (type I errors): 38.1 -> 15.9Improved
Use early dropping with K=infinityFBS · Null data, p=200, alpha=0.05number of selected variables (type I errors): 20.8 -> 22.3No clear effect
Use early dropping with K=0 or 1FBS · 12 binary datasets, same hyper-parametersnumber of independence tests: around 1-2 orders of magnitude fewer testsImproved
Use early dropping with K=infinityFBS · 12 binary datasets, same hyper-parametersnumber of independence tests: 25%-30% of the testsImproved
Use early dropping with K=0 or 1FBS · 12 binary datasets, same hyper-parametersAUC / accuracyNo clear effect
Use early dropping with K=0 (IT)FBS (IT) · muskAUC: 96.9 -> 93.0Hurt
Use early dropping with K=0 (EBIC)FBS (EBIC) · 12 binary datasets, optimized hyper-parametersnumber of selected variables: avg rank 3.21 -> 1.17Improved
Use independence test instead of EBICFBS-ED K=0 · 12 binary datasets, optimized hyper-parametersAUC average rank: 6.33 -> 3.75Improved
Use LASSO-FS instead of FBS-ED K=0 (IT)FBS-ED K=0 (IT) · 12 binary datasets, optimized hyper-parametersAUC average rank: 3.75 -> 1.42Improved
Use LASSO-FS instead of FBS-ED K=0 (IT)FBS-ED K=0 (IT) · arceneAUC: 78.6 -> 83.7Improved
Limit LASSO-FS to the same number of variables as forward selectionLASSO-FS · 12 binary datasetsAUC / accuracy differenceNo clear effect

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy