Memorization and generalization in generative diffusion under the manifold hypothesis

DOI10.1088/1742-5468/ADE136MaRDI QIDQ7234619zbMATH OpenFDO

Authors Beatrice Achilli, Luca Ambrogioni, Carlo Lucibello, Marc Mézard, Enrico Ventura

Publication date 28 April 2026

Published in Journal of Statistical Mechanics: Theory and Experiment (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1088/1742-5468/ADE136

zbMATH Keywords

cavity and replica methodmachine learningstatistical inferencestochastic processes

Mathematics Subject Classification ID

Statistical mechanics, structure of matter (82-XX)

Cites work

• A phase transition in diffusion models reveals the hierarchical nature of data
• Approximation by superpositions of a sigmoidal function
• Error bounds for approximations with deep ReLU networks
• Generalization error of GAN from the discriminator's perspective
• Generative diffusion in very large dimensions
• On a model of associative memory with huge storage capacity
• Random-energy model: an exactly solvable model of disordered systems

Cited in

(1)

• Geometric regularity in deterministic sampling dynamics of diffusion-based generative models

This page was built for publication: Memorization and generalization in generative diffusion under the manifold hypothesis

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7234619)