AI Paper Atlas

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

GULP: a prediction-based metric between representations

Enric Boix-Adserà, Hannah Lawrence, George Stepaniants, Philippe Rigollet

NeurIPS 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory · 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
Measure representation distance using uniform prediction differences over regularized linear tasksImageNet pretrained representations · ImageNet; linear regressionSpearman's ρ with prediction differencesImproved
Use GULP with larger regularizationImageNet pretrained representations · ImageNet architecture clusteringarchitecture separationImproved
Increase GULP regularizationIndependently trained ResNet18 representations · CIFAR10 trainingdistance trend across trainingImproved
Use GULP to predict generalization differences for logistic rather than linear predictionGULP · Downstream logistic predictionprediction-gap correlationHurt
Set λ=0, recovering the CCA-based distance (ablation)GULP · Representation comparisondistance invarianceMixed
Set λ=10⁻⁶ rather than using larger regularization (ablation)GULP on MNIST network representations · MNIST trainarchitecture clusteringMixed
Use GULP with λ=1 rather than λ=10⁻⁶ (ablation)GULP on MNIST network representations · MNIST trainarchitecture clusteringImproved
Use small λ rather than large λ during training (ablation)GULP on independently trained ResNet18 representations · CIFAR10 trainingsensitivity to representation differencesMixed

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