G
Geert-Jan Huizing
Researcher at École Normale Supérieure
Publications - 1
Citations - 2
Geert-Jan Huizing is an academic researcher from École Normale Supérieure. The author has contributed to research in topics: Eigenvalues and eigenvectors & Metric (unit). The author has an hindex of 1, co-authored 1 publications receiving 2 citations.
Papers
More filters
Posted Content
Unsupervised Ground Metric Learning using Wasserstein Eigenvectors.
TL;DR: In this article, the ground cost is computed as a positive eigenvector of the function mapping a cost to the pairwise optimal transport distances between the inputs, which is homogeneous and monotone, thus framing unsupervised metric learning as a nonlinear Perron-Frobenius problem.