Saved in:
Bibliographic Details
Main Authors: Williams, Bernardo, Yeom-Song, Victor M., Hartmann, Marcelo, Klami, Arto
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.27480
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918357909897216
author Williams, Bernardo
Yeom-Song, Victor M.
Hartmann, Marcelo
Klami, Arto
author_facet Williams, Bernardo
Yeom-Song, Victor M.
Hartmann, Marcelo
Klami, Arto
contents We propose a method for learning and sampling from probability distributions supported on the simplex. Our approach maps the open simplex to Euclidean space via smooth bijections, leveraging the Aitchison geometry to define the mappings, and supports modeling categorical data by a Dirichlet interpolation that dequantizes discrete observations into continuous ones. This enables density modeling in Euclidean space through the bijection while still allowing exact recovery of the original discrete distribution. Compared to previous methods that operate on the simplex using Riemannian geometry or custom noise processes, our approach works in Euclidean space while respecting the Aitchison geometry, and achieves competitive performance on both synthetic and real-world data sets.
format Preprint
id arxiv_https___arxiv_org_abs_2510_27480
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Simplex-to-Euclidean Bijections for Categorical Flow Matching
Williams, Bernardo
Yeom-Song, Victor M.
Hartmann, Marcelo
Klami, Arto
Machine Learning
We propose a method for learning and sampling from probability distributions supported on the simplex. Our approach maps the open simplex to Euclidean space via smooth bijections, leveraging the Aitchison geometry to define the mappings, and supports modeling categorical data by a Dirichlet interpolation that dequantizes discrete observations into continuous ones. This enables density modeling in Euclidean space through the bijection while still allowing exact recovery of the original discrete distribution. Compared to previous methods that operate on the simplex using Riemannian geometry or custom noise processes, our approach works in Euclidean space while respecting the Aitchison geometry, and achieves competitive performance on both synthetic and real-world data sets.
title Simplex-to-Euclidean Bijections for Categorical Flow Matching
topic Machine Learning
url https://arxiv.org/abs/2510.27480