New Rate-Distortion-Perception Theory Redefines Information Limits

A new theory, Rate-Distortion-Perception (RDP) theory, has been proposed to redefine the fundamental limits of information representation. This theory extends classical rate-distortion theory by incorporating perception as a third crucial axis, quantified through distributional similarity between original and reconstructed signals. The RDP framework aims to better capture perceptual quality and semantic validity, which are increasingly important in modern AI-driven applications, by utilizing various perceptual constraints like f-divergences and Wasserstein-based metrics. AI

IMPACT This new theory could lead to more efficient and semantically aware data compression techniques, impacting AI model training and deployment.

RANK_REASON The item is a research paper detailing a new theoretical framework. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX Code Finder for Papers
• Connected Papers
• DagsHub
• Gotit.pub
• Hugging Face
• Litmaps
• Photios A. Stavrou
• Rate-Distortion-Perception Theory
• ScienceCast
• scite Smart Citations

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →