AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Supervising Unsupervised Learning

Vikas Garg, Adam T Kalai

NeurIPS 2018 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Zero-shot robustness and transfer · 24 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 estimated ARI to choose k instead of maximizing Silhouette scoreK-means on OpenML clustering problems · OpenML clustering problemsARIImproved
Use estimated ARI to choose k instead of maximizing Silhouette scoreK-means on OpenML clustering problems · OpenML clustering problemsRMSEImproved
Select the clustering algorithm using learned problem and cluster featuresClustering on OpenML binary classification problems · OpenML binary classification problemsARI: by over 5%Improved
Remove 1% of instances as outliers before clusteringClustering on OpenML classification problems · OpenML classification problemsARIImproved
Remove more than 2% of instances as outliersClustering on OpenML classification problems · OpenML classification problemsARI: beyond 2%Hurt
Learn a binary similarity function from archived supervised datasetsPairwise clustering similarity prediction · Meta-ITaccuracyImproved
Learn a binary similarity function from archived supervised datasets, excluding the test category from trainingPairwise clustering similarity prediction · Meta-ETaccuracyImproved

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