AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

A Scalable and Efficient Iterative Method for Copying Machine Learning Classifiers

Nahuel Statuto, Irene Unceta, Jordi Nin, Oriol Pujol

JMLR 2023 · Probabilistic and scientific modeling › Molecules and proteins · approach: Prediction-powered and conformal inference · 0 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 sequential copying instead of single-pass copying (Best accuracy point)single-pass copy · 58 UCI datasetscopy accuracy: matches in 53 of 58, worse in 4, better in 1No clear effect
Use sequential copying (Best accuracy point) instead of single-pass copyingsingle-pass copy · 58 UCI datasetssamples used: 14% of single-pass samplesImproved
Choose Best efficiency operating point over Best accuracysequential copy · 58 UCI datasetsefficiency eff: 0.716 -> 0.882Improved
Choose Best efficiency operating point over Best accuracysequential copy · 58 UCI datasetsaccuracy: 5% average degradationHurt
Choose Best efficiency operating point over Best accuracysequential copy · 58 UCI datasetsconvergence time: 11.5% -> 17.4% of allotted timeMixed
Choose Best convergence operating point over Best accuracysequential copy · 58 UCI datasetsconvergence time: 11.5% -> 11.2% of allotted timeNo clear effect
Add automatic lambda (start large, decrease) (ablation)sequential copy · 58 UCI datasetsconvergence and accuracyImproved
Add sample removal by uncertainty threshold delta (ablation)sequential copy · 58 UCI datasetstraining samples used: eff 0.716 (Best accuracy)Improved

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