AI Paper Atlas

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

Attraction-Repulsion Spectrum in Neighbor Embeddings

Jan Niklas Böhm, Philipp Berens, Dmitry Kobak

NeurIPS 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning

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
Increase t-SNE attraction by raising exaggeration from ρ=1 to ρ=4t-SNE · MNISTkNN recall: 0.34 -> 0.12Hurt
Increase t-SNE attraction by raising exaggeration from ρ=4 to ρ=30t-SNE · MNISTkNN recall: 0.12 -> 0.06Hurt
Increase t-SNE attractiont-SNE · Simulated 20-Gaussian trajectorycontinuity of the recovered manifoldImproved
Increase t-SNE attractiont-SNE · Human brain organoid developmental datarepresentation of developmental trajectoryImproved
Increase t-SNE repulsion by using standard exaggeration instead of strong attractiont-SNE · Human brain organoid developmental datarepresentation of cluster structureImproved
Use UMAP negative sampling with larger rate m (ablation)UMAP · MNIST subsamplerepulsion strengthImproved
Increase negative sampling rate m while decreasing γ to keep γ·m constant (ablation)UMAP · MNIST subsampleembeddingNo clear effect
Decrease UMAP repulsion coefficient γ (ablation)UMAP · MNISTattraction-repulsion spectrum positionImproved
Set UMAP output-kernel parameters to a=b=1 (ablation)UMAP · MNISTembeddingNo clear effect
Set UMAP ε=1 (ablation)UMAP · MNISTembeddingNo clear effect
Remove degree-based repulsion (ablation)ForceAtlas2 · MNISTembedding scale: ∼15 times decrease in scaleHurt
Remove degree-based repulsion (ablation)ForceAtlas2 · MNISTembedding layoutNo clear effect

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