AI Paper Atlas

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

Generalized Resubstitution for Classification Error Estimation

Parisa Ghane, Ulisses Braga-Neto

JMLR 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Group-robust classification · 2 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 3NN posterior-probability estimator instead of zero bootstraperror estimation, linear SVM · Synthetic data, n=20computation time (ms): 138.75 -> 2.31Improved
use spherical bolstered resubstitution instead of zero bootstraperror estimation, RBF SVM · Synthetic data, n=100computation time (ms): 434.10 -> 61.53Improved
use 10-fold cross-validation instead of spherical bolstered resubstitutionerror estimation, 3-NN · Synthetic data, n=100computation time (ms): 343.50 -> 13.11Improved
add bolstering kernels to 3NN posterior-probability estimator3NNpp, 3-NN classifier · Synthetic data, n=20computation time (ms): 4.00 -> 136.46Hurt
increase sample size from n=20 to n=100 for spherical bolstered resubstitutionspherical bolstered resubstitution, linear SVM · Synthetic datacomputation time (ms): 9.65 -> 41.06Hurt
replace plain resubstitution with bolstered resubstitutionresubstitution error estimator · Synthetic data, linear and CART classifiersbias and RMS of error estimateImproved
use bolstered resubstitution on 3NN classifierbolstered resubstitution · Synthetic data, 3NN classifierbias of error estimateNo clear effect
use semi-bolstered resubstitution instead of bolstered resubstitution3NN classifier error estimation · Synthetic data, 3NN classifierRMS of error estimateImproved
use bolstered posterior-probability estimator instead of cross-validation and bootstraperror estimation · Synthetic data, non-3NN classifiersRMS of error estimateImproved
use Naive-Bayes bolstered resubstitution instead of plain resubstitutionLeNet-5 error estimation · MNIST subsets n=200-800bias of error estimateImproved
use 11-NN posterior-probability and bolstered posterior-probability estimators instead of plain resubstitutionLeNet-5 error estimation · MNIST subsets n=200-800bias of error estimateHurt
add data-driven kappa calibration to Naive-Bayes bolstered resubstitutionNaive-Bayes bolstered resubstitution · MNIST subsets n=200-800bias and RMS of error estimateImproved

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