AI Paper Atlas

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

Getting Better from Worse: Augmented Bagging and A Cautionary Tale of Variable Importance

Lucas Mentch, Siyu Zhou

JMLR 2022 · Learning theory and algorithms › Prediction and decision losses · approach: Provable post-hoc explanations · 14 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
add independent noise features to bagging (AugBagg)Bagging with random forest comparison · Linear simulation, SNR 0.01relative test error: surpasses the best random forest at around q=25Improved
add independent noise features to bagging (AugBagg)AugBagg · Linear simulation, SNR 0.14relative test error: helps until approximately q=50, then deterioratesMixed
make noise features correlated with an original feature (ablation)AugBagg · Linear simulation, SNR 0.01 and 0.05relative test error: all AugBagg models outperform best random forest once q>50, except r=0.99Improved
add noise features correlated with original features (ablation)AugBagg · Linear simulation, SNR 1.22 and 2.07relative test error: AB(q,0.99) is the only model not deteriorating substantially with qHurt
tuned AugBagg instead of bagging on original dataBagging · 14 real datasets with injected response noiserelative test error: outperforms bagging on 12 of 14 datasetsImproved
add noise features on high-dimensional datasets AquaticTox and mtp2Bagging · AquaticTox and mtp2relative test error: bagging remains superior; lasso selects 17% and 4% of featuresHurt
drop the tested noise features instead of replacing them (ablation)Variable importance test · Independent noise features, low SNR, large qrejection probability: rejects nearly 30% versus nominal 5%Hurt
use replacement tests with matching noise distribution (ablation)Variable importance test · Replacement tests, same distributionrejection probability: rejection rate stays near nominal 5%Improved
replace correlated noise (r=0.7) with independent noise in replacement test (ablation)Variable importance test · Mismatched replacement distributionrejection probability: very high rejection probability across all but lowest SNRs, rising with qHurt

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