AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

A Heat Diffusion Perspective on Geodesic Preserving Dimensionality Reduction

Guillaume Huguet, Alexander Tong, Edward De Brouwer, Yanlei Zhang, Guy Wolf, Ian Adelstein, Smita Krishnaswamy

NeurIPS 2023 · Image generation › Fast diffusion sampling · approach: Diffusion models as perception tools · 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
Use HeatGeo instead of shortest-path geodesic estimateHeatGeo · Swiss roll (noise 1.0)Pearson correlation: 0.497 ± 0.144 -> 0.702 ± 0.086Improved
Use HeatGeo instead of PHATE for distance recoveryHeatGeo · TreePearson correlation: 0.766 ± 0.023 -> 0.822 ± 0.008Improved
Use HeatGeo instead of PHATE for clusteringHeatGeo · PBMChomogeneity: 0.798 ± 0.012 -> 0.734 ± 0.009Hurt
Use HeatGeo instead of UMAP for clusteringHeatGeo · PBMChomogeneity: 0.177 ± 0.037 -> 0.734 ± 0.009Improved
Use HeatGeo instead of UMAP for clusteringHeatGeo · TreeaMI: 0.681 ± 0.086 -> 0.786 ± 0.051Improved
Use HeatGeo instead of UMAP for clusteringHeatGeo · Swiss rollhomogeneity: 0.810 ± 0.036 -> 0.820 ± 0.008No clear effect
Use HeatGeo instead of UMAP for held-out time point interpolationHeatGeo · IPSCEMD: 0.866 ± 0.058 -> 0.365 ± 0.056Improved
Use HeatGeo instead of UMAP for held-out time point interpolationHeatGeo · CiteEMD: 0.791 ± 0.045 -> 0.890 ± 0.046Hurt
Use HeatGeo instead of t-SNE for held-out time point interpolationHeatGeo · MultiEMD: 1.208 ± 0.087 -> 0.958 ± 0.044Improved
Use HeatGeo instead of Diffusion Maps for clusteringHeatGeo · Treehomogeneity: 0.341 ± 0.103 -> 0.784 ± 0.051Improved
Increase noise level from 0.1 to 1.0 for Heat-Geo distance recoveryHeat-Geo · Swiss rollPearsonR: 0.992 ± 0.003 -> 0.702 ± 0.086Hurt
Increase noise level from 0.1 to 1.0 for PHATE distance recoveryPHATE · Swiss rollPearsonR: 0.621 ± 0.006 -> 0.457 ± 0.01Hurt

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