AI Paper Atlas

Research map / Learning theory and algorithms / Optimal transport

Computing Kantorovich-Wasserstein Distances on $d$-dimensional histograms using $(d+1)$-partite graphs

Gennaro Auricchio, Federico Bassetti, Stefano Gualandi, Marco Veneroni

NeurIPS 2018 · Learning theory and algorithms › Optimal transport · approach: Wasserstein barycenters and GW · 5 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 the 3-partite graph instead of the bipartite graphExact optimal transport with squared ℓ2 cost · DOTmark 32×32, Classicruntime (seconds): 0.54 (0.05) -> 0.07 (0.01)Improved
Use the 3-partite graph instead of the bipartite graphExact optimal transport with squared ℓ2 cost · DOTmark 128×128, Classicruntime (seconds): 1\,368 (545) -> 36.2 (5.4)Improved
Use the (d+1)-partite graph instead of the bipartite graphExact optimal transport with separable cost · Flow Cytometry AML, N=16, d=3runtime (seconds): 38.2 (14.0) -> 0.12 (0.02)Improved
Use the (d+1)-partite graph instead of the bipartite graphExact optimal transport with separable cost · Flow Cytometry AML, N=16, d=4runtime: out-of-memory -> 4.8 (0.84)Improved
Use the (d+1)-partite graph instead of the bipartite graphExact optimal transport with separable cost · Flow Cytometry AML, N=32, d=3runtime: out-of-memory -> 5.23 (0.69)Improved
Use the 3-partite exact method instead of Improved Sinkhorn λ=1.5Optimal transport on regular 2D histograms · DOTmark 32×32, Classicoptimality gap: 9.1% -> 0Improved
Use Improved Sinkhorn λ=1.5 instead of the exact methodSinkhorn optimal transport · DOTmark 32×32, GRFroughoptimality gap: 0 -> 31.0%Hurt
Use Improved Sinkhorn λ=1.5 instead of the exact methodSinkhorn optimal transport · DOTmark 32×32, WhiteNoiseoptimality gap: 0 -> 39.2%Hurt

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