AI Paper Atlas

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

Steering Distortions to Preserve Classes and Neighbors in Supervised Dimensionality Reduction

Benoît Colange, Jaakko Peltonen, Michael Aupetit, Denys Dutykh, Sylvain Lespinats

NeurIPS 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Data augmentation and generalization bias · 4 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
add class supervision (eps 0 to 0.5) to NeRVNeRV · Isolet 510-NN accuracy (%): 71.0 -> 91.9Improved
increase supervision level eps (ablation)ClassNeRV · Globe10-NN accuracyImproved
raise tau* toward 1 (ablation)ClassNeRV · Globetrustworthiness vs continuityMixed
use ClassNeRV instead of NCAClassNeRV · Isolet 5neighborhood structure preservationImproved
use ClassNeRV instead of tSNEClassNeRV · Isolet 5trustworthinessHurt
use true labels instead of random labels (ablation)ClassNeRV · Digitsclass preservationImproved

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