AI Paper Atlas

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

DeltaGrad: Rapid retraining of machine learning models

Yinjun Wu, Edgar Dobriban, Susan Davidson

ICML 2020 · Learning theory and algorithms › Prediction and decision losses · approach: Active learning and subset selection · 42 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 DeltaGrad instead of retraining from scratch for batch addition/deletionRegularized logistic regression and two-layer neural network · MNIST, Covertype, HIGGS, RCV1accuracy: BaseL and DeltaGrad accuracies are equal or differ only slightly; confidence intervals overlapNo clear effect
Use DeltaGrad instead of retraining from scratch for batch addition/deletionRegularized logistic regression · MNISTspeedup: up to 2.6xImproved
Use DeltaGrad instead of retraining from scratch for batch addition/deletionRegularized logistic regression · Covertypespeedup: 2xImproved
Use DeltaGrad instead of retraining from scratch for batch addition/deletionRegularized logistic regression · HIGGSspeedup: 1.6xImproved
Use DeltaGrad instead of retraining from scratch for batch addition/deletionRegularized logistic regression · RCV1speedup: 6.5xImproved
Use DeltaGrad instead of retraining from scratch for online updatesLogistic regression · MNIST, Covertype, HIGGS, RCV1accuracy: BaseL and DeltaGrad prediction accuracies are identical in Table 2No clear effect
Use DeltaGrad instead of retraining from scratch for online updatesLogistic regression · MNISTspeedup: 2.5xImproved
Use DeltaGrad instead of retraining from scratch for online updatesLogistic regression · Covertypespeedup: 2xImproved
Use DeltaGrad instead of retraining from scratch for online updatesLogistic regression · HIGGSspeedup: 1.8xImproved
Use DeltaGrad instead of retraining from scratch for online updatesLogistic regression · RCV1speedup: 6.5xImproved
Use the non-strongly-convex DeltaGrad variant instead of retraining from scratchTwo-layer neural network · MNISTspeedup: about 1.4xImproved

Later papers that compared with this method

1 later accepted paper reports DeltaGrad as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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