Rate-Distortion-Perception Theory for the Quadratic Wasserstein Space

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Qu, Xiqiang, Chen, Jun, Yu, Lei, Xu, Xiangyu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908335902556160
author Qu, Xiqiang
Chen, Jun
Yu, Lei
Xu, Xiangyu
author_facet Qu, Xiqiang
Chen, Jun
Yu, Lei
Xu, Xiangyu
contents We establish a single-letter characterization of the fundamental distortion-rate-perception tradeoff with limited common randomness under the squared error distortion measure and the squared Wasserstein-2 perception measure. Moreover, it is shown that this single-letter characterization can be explicitly evaluated for the Gaussian source. Various notions of universal representation are also clarified.
format Preprint
id arxiv_https___arxiv_org_abs_2504_17236
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Rate-Distortion-Perception Theory for the Quadratic Wasserstein Space
Qu, Xiqiang
Chen, Jun
Yu, Lei
Xu, Xiangyu
Information Theory
Machine Learning
We establish a single-letter characterization of the fundamental distortion-rate-perception tradeoff with limited common randomness under the squared error distortion measure and the squared Wasserstein-2 perception measure. Moreover, it is shown that this single-letter characterization can be explicitly evaluated for the Gaussian source. Various notions of universal representation are also clarified.
title Rate-Distortion-Perception Theory for the Quadratic Wasserstein Space
topic Information Theory
Machine Learning
url https://arxiv.org/abs/2504.17236