AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

On the Stability of Feature Selection Algorithms

Sarah Nogueira, Konstantinos Sechidis, Gavin Brown

JMLR 2018 · Label-efficient and robust learning › Recognition and long-tail · approach: Data augmentation and generalization bias · 186 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
Choose LASSO lambda by maximizing stability along with likelihood instead of likelihood onlyL1-logistic regression · Synthetic data, low redundancyfalse positives: FP = 14 -> FP = 0 (FN 19 -> 20)Improved
Choose LASSO lambda by likelihood and stability instead of likelihood onlyL1-logistic regression · Synthetic data, high redundancyfalse positives: FP = 12 -> FP = 0 (FN 38 -> 39)Improved
Choose lambda by likelihood and stability instead of likelihood onlyL1-logistic regression · Synthetic data, medium redundancyfalse positives / false negatives: FP = 0, FN = 26 -> FP = 0, FN = 26No clear effect
Choose lambda by likelihood and stability instead of likelihood onlyL1-logistic regression · Synthetic data, no redundancyfalse positives / false negatives: FP = 0, FN = 0 -> IDEMNo clear effect
Increase LASSO lambda from 0.0051 (likelihood optimum) to 0.0187 (ablation)L1-logistic regression · Synthetic data, low redundancystability: stability up to 0.63 with negligible likelihood lossImproved
Pick lambda to optimize stability as well as likelihoodL1-logistic regression · Synthetic data, all redundancy levelsmisclassification errorNo clear effect
Pick Elastic Net lambda 0.76 instead of likelihood-optimal 0.05 (ablation)Elastic Net alpha=0.05 · Synthetic data, high redundancystability: 0.34 -> 0.98Improved
Pick Elastic Net lambda 0.76 instead of 0.05 (ablation)Elastic Net alpha=0.05 · Synthetic data, high redundancyOOB misclassification error: 14% -> 14%No clear effect
Use Elastic Net (L1+L2) instead of LASSO under high redundancyRegularized logistic regression · Synthetic data, high redundancystability: stability greater than 0.90 for lambda > 0.56Improved
Apply Stability Selection on top of LASSOLASSO · Spambase, Boston housing, Sonar, MadelonstabilityImproved
Stability Selection vs LASSO, significance test at 5%Stability Selection · Sonar and Madelonstability significanceMixed
Increase number of samples MStability estimator · Toy Bernoulli data d=100estimation error |Phi-hat - Phi|: within 5% for M <= 10, within 1% for M <= 100Improved

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