Learning Diffusion Priors from Observations by Expectation Maximization

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Learning Diffusion Priors from Observations by Expectation Maximization
Συγγραφείς: Rozet, François, Andry, Gérôme, Lanusse, François, Louppe, Gilles
Πηγή: Advances in Neural Information Processing Systems, 37 (2024-05-22); The Thirty-Eighth Annual Conference on Neural Information Processing Systems, Vancouver, Canada [CA], December 10-15, 2024
Στοιχεία εκδότη: Curran Associates, 2024.
Έτος έκδοσης: 2024
Θεματικοί όροι: Computer Science - Learning, Statistics - Machine Learning, Engineering, computing & technology, Computer science, Ingénierie, informatique & technologie, Sciences informatiques
Περιγραφή: Diffusion models recently proved to be remarkable priors for Bayesian inverse problems. However, training these models typically requires access to large amounts of clean data, which could prove difficult in some settings. In this work, we present DiEM, a novel method based on the expectation-maximization algorithm for training diffusion models from incomplete and noisy observations only. Unlike previous works, DiEM leads to proper diffusion models, which is crucial for downstream tasks. As part of our methods, we propose and motivate an improved posterior sampling scheme for unconditional diffusion models. We present empirical evidence supporting the effectiveness of our approach.
Τύπος εγγράφου: conference paper
http://purl.org/coar/resource_type/c_5794
conferenceObject
peer reviewed
Γλώσσα: English
Relation: https://openreview.net/forum?id=7v88Fh6iSM; https://openreview.net/forum?id=7v88Fh6iSM; urn:issn:1049-5258
Σύνδεσμος πρόσβασης: https://orbi.uliege.be/handle/2268/319891
Rights: open access
http://purl.org/coar/access_right/c_abf2
info:eu-repo/semantics/openAccess
Αριθμός Καταχώρησης: edsorb.319891
Βάση Δεδομένων: ORBi
Περιγραφή
Η περιγραφή δεν είναι διαθέσιμη