AI Paper Atlas

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

AutoCoreset: An Automatic Practical Coreset Construction Framework

Alaa Maalouf, Murad Tukan, Vladimir Braverman, Daniela Rus

ICML 2023 · Learning theory and algorithms › Prediction and decision losses · approach: Loss landscape analysis · 1 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 AutoCoreset instead of competing coreset methodslogistic regression · real-world classification datasetstest accuracy: 5%Improved
use AutoCoreset instead of competing coreset methodslogistic regression · real-world classification datasetsapproximation error: 30Improved
use AutoCoreset instead of competing coreset methodsSVM · real-world classification datasetstest accuracyImproved
use AutoCoreset instead of competing coreset methodsSVM · real-world classification datasetsapproximation errorImproved
use AutoCoreset instead of competing coreset methodslinear regression · real-world regression datasetsapproximation errorImproved
use AutoCoreset instead of competing coreset methodslinear regression · real-world regression datasetsR^2Improved
use AutoCoreset instead of competing coreset methodsk-means · real-world clustering datasetsapproximation errorImproved
use AutoCoreset instead of uniform samplingAutoCoreset · CIFAR10 and TinyImageNetrelative approximation error: ≈ 1.5Improved
select the coreset with the best observed cost instead of returning the last coreset (ablation)AutoCoreset · real-world experimentsapproximation errorImproved
return the last coreset instead of selecting the coreset with the best observed cost (ablation)AutoCoreset · some real-world experimentsapproximation error: 2 to 3 samplesHurt

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