Dequantified Diffusion-Schrödinger Bridge for Density Ratio Estimation

Wei Chen, Shigui Li, Jiacheng Li, Junmei Yang, John Paisley, Delu Zeng · 2025

Read Dequantified Diffusion-Schrödinger Bridge for Density Ratio Estimation by Wei Chen, Shigui Li, Jiacheng Li, Junmei Yang, John Paisley, Delu Zeng (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-05034.txt.

Description

Wei Chen and co-authors improve a basic statistical tool that compares two probability distributions, in work from 2025. Estimating the ratio between two densities feeds many tasks, but methods break down when the distributions barely overlap.

Two failure modes have names: the density chasm and the support chasm. Scores also blow up near boundaries, making estimates unstable. The team's framework builds a smoothed bridge between the distributions using diffusion steps and added Gaussian noise, which widens support and calms the scores. A second version adds optimal transport for extra accuracy. Tests on information estimation and density tasks beat prior baselines.

Catalog record

CREATORWei Chen, Shigui Li, Jiacheng Li, Junmei Yang, John Paisley, Delu Zeng
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE100 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-05034

Browse the full library · About the Tower of Babel