George Zhang

2 accepted papers

2022

LOSSY COMPRESSION WITH DISTRIBUTION SHIFT AS ENTROPY CONSTRAINED OPTIMAL TRANSPORT

Huan Liu, George Zhang, Jun Chen, Ashish J Khisti

ICLR 2022poster

We study an extension of lossy compression where the reconstruction distribution is different from the source distribution in order to account for distributional shift due to processing. We formulate this as a generalization of optimal transport with an entropy bottleneck to account for the rate con…

Cited by 15 Source ↗PDF ↗Scholar ↗

2021

Universal Rate-Distortion-Perception Representations for Lossy Compression

George Zhang, Jingjing Qian, Jun Chen, Ashish J Khisti

NeurIPS 2021poster

In the context of lossy compression, Blau \& Michaeli (2019) adopt a mathematical notion of perceptual quality and define the information rate-distortion-perception function, generalizing the classical rate-distortion tradeoff. We consider the notion of universal representations in which one may fix…

Cited by 70 Source ↗PDF ↗Scholar ↗