AI Paper Atlas

Research map / Learning theory and algorithms / Sample complexity

V-statistics and Variance Estimation

Zhengze Zhou, Lucas Mentch, Giles Hooker

JMLR 2021 · Learning theory and algorithms › Sample complexity · approach: Optimal mean estimation and subsampling · 17 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 bias correction to the variance estimator (corrected-V instead of IJ) (ablation)random forest variance estimation · MARS, p1, k=100, B=500variance ratio (ideal 1): 7.6985 -> 1.4962Improved
add bias correction to the variance estimator (ablation)random forest variance estimation · MARS, p1, k=100, B=5000variance ratio (ideal 1): 1.7073 -> 1.0305Improved
add bias correction to the variance estimator (ablation)random forest variance estimation · MARS, p1, k=100, B=250095% CI coverage (%): 99.2 -> 94.2Improved
add bias correction to the variance estimator (ablation)random forest variance estimation · MARS, p1, k=500, B=500variance ratio (ideal 1): 10.7237 -> 4.5438Improved
increase number of trees B from 500 to 5000IJ original variance estimate · MARS, p1, k=100variance ratio (ideal 1): 7.6985 -> 1.7073Improved
increase subsample size k from 100 to 500corrected-V variance estimate · MARS, p1, B=500variance ratio (ideal 1): 1.4962 -> 4.5438Hurt
sample with replacement (V) instead of without (U), full subsample sizerandom forest · diabetesRMSE: 58.7494 -> 55.4130Improved
sample with replacement (V) instead of without (U), full subsample sizerandom forest · bostonRMSE: 3.0588 -> 3.2944Hurt
reduce subsample proportion from 1 to 0.2 (U-statistics)random forest · diabetesRMSE: 58.7494 -> 54.6137Improved
reduce subsample proportion from 1 to 0.2 (U-statistics)random forest · bostonRMSE: 3.0588 -> 3.9696Hurt
use Balanced Variance Estimation instead of Internal Variance Estimationvariance estimator · simulation, n=500, k=100variance biasImproved
apply bias correction to U-statistics ensemblescorrected estimator · sampling without replacementvariance estimateHurt

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