AI Paper Atlas

Research map / Optimization / Nonconvex and smooth optimization

Data Summarization via Bilevel Optimization

Zalán Borsos, Mojmír Mutný, Marco Tagliasacchi, Andreas Krause

JMLR 2024 · Optimization › Nonconvex and smooth optimization · approach: Penalty and implicit-gradient bilevel methods

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
build coreset jointly for WideResNet-16-4 and VGG16 (ablation)BiCo · CIFAR-10, VGG16test accuracy: 93.66 -> 93.88Improved
build coreset jointly for WideResNet-16-4 and VGG16 (ablation)BiCo · CIFAR-10, MobileNetV2 (transfer)test accuracy: 93.65 -> 93.79Improved
build coreset jointly for WideResNet-16-4 and VGG16 (ablation)BiCo · SVHN, VGG16test accuracy: 95.43 -> 95.75Improved
build coreset jointly for WideResNet-16-4 and VGG16 (ablation)BiCo · SVHN, MobileNetV2 (transfer)test accuracy: 95.53 -> 95.76Improved
build coreset jointly for WideResNet-16-4 and VGG16 (ablation)BiCo · CIFAR-10, WideResNet-16-4test accuracy: 95.24 -> 95.18Hurt
build coreset jointly for WideResNet-16-4 and VGG16 (ablation)BiCo · SVHN, WideResNet-16-4test accuracy: 96.97 -> 96.88Hurt
use BiCo coreset instead of uniform sampling (transferred)WideResNet-16-4 coreset size 23000 · CIFAR-10, VGG16test accuracy: 91.49 -> 93.66Improved
use BiCo coreset instead of uniform sampling (transferred)WideResNet-16-4 coreset size 23000 · SVHN, MobileNetV2test accuracy: 94.22 -> 95.53Improved
use BiCo replay memory instead of uniform samplingcontinual learning · SMNIST, memory 50accuracy: 85.23 -> 91.61Improved
use BiCo replay memory instead of uniform samplingcontinual learning · SMNIST, memory 200accuracy: 95.08 -> 97.01Improved
use BiCo merge-reduce buffer instead of reservoir samplingstreaming · SMNIST, memory 50accuracy: 83.90 -> 85.32Improved
add sample weights (regularization-based weighted coreset) to unweighted selectionBiCo on binary logistic regression · LIBSVM binary datasetscoreset size to match full-data accuracy: 2% of the data vs twice as largeImproved

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