AI Paper Atlas

Research map / Learning theory and algorithms / Sample complexity

More Efficient Estimation for Logistic Regression with Optimal Subsamples

HaiYing Wang

JMLR 2019 · Learning theory and algorithms › Sample complexity · approach: Optimal mean estimation and subsampling

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 unweighted estimator combining pilot and second-step estimates instead of weighted OSMACOSMAC logistic regression · simulation, six covariate distributionsrelative efficiency (MSE ratio)Improved
use unweighted estimator instead of weighted with LoptOSMAC logistic regression · simulation T3 covariatesrelative efficiencyMixed
Poisson subsampling instead of subsampling with replacementunweighted OSMAC estimator · simulation, six covariate distributionsrelative efficiency: up to 250% as efficient as weighted estimator (nzNormal, Aopt)Improved
use Aopt probabilities instead of LoptPoisson subsampling · SUSY, n0=200relative efficiencyImproved
use optimal probabilities instead of uniform subsamplingsubsampling estimator · simulationempirical MSEImproved
increase second-step subsample size n with Loptunweighted/Poisson estimators · SUSY, n0=200relative efficiencyImproved
Poisson subsampling instead of replacement, data on hard driveOSMAC · CPU time, N=10^7, LoptCPU seconds: 4384.99 -> 2699.13Improved
unweighted combination instead of weighted, data in RAMLopt, N=10^7 · CPU time in RAMCPU seconds: 5.24 -> 3.71Improved
Aopt instead of Lopt probabilities, data on hard driveOSMAC · CPU timeCPU seconds: 4384.99 -> 4369.04No clear effect
Aopt instead of Lopt probabilities, data in RAMOSMAC · CPU time, N=10^7CPU seconds: 3.71 -> 34.98Hurt
subsampling instead of full dataPoisson subsampling · CPU time on hard drive, N=10^7CPU seconds: 138134.69 -> 2699.13Improved

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