Self-Consuming Generative Models Go MAD

Simple item page
dc.contributor.advisorBaraniuk, Richard Gen_US
dc.creatorCasco-Rodriguez, Josueen_US
dc.date.accessioned2024-08-30T16:42:40Zen_US
dc.date.available2024-08-30T16:42:40Zen_US
dc.date.created2024-08en_US
dc.date.issued2024-07-30en_US
dc.date.submittedAugust 2024en_US
dc.date.updated2024-08-30T16:42:40Zen_US
dc.description.abstractSeismic advances in generative AI algorithms for imagery, text, and other data types has led to the temptation to use synthetic data to train next-generation models. Repeating this process creates an autophagous (self-consuming) loop whose properties are poorly understood. We conduct a thorough analytical and empirical analysis using state-of-the-art generative image models of three families of autophagous loops that differ in how fixed or fresh real training data is available through the generations of training and in whether the samples from previous generation models have been biased to trade off data quality versus diversity. Our primary conclusion across all scenarios is that without enough fresh real data in each generation of an autophagous loop, future generative models are doomed to have their quality (precision) or diversity (recall) progressively decrease. We term this condition Model Autophagy Disorder (MAD), making analogy to mad cow disease.en_US
dc.format.mimetypeapplication/pdfen_US
dc.identifier.citationCasco-Rodriguez, Josue. Self-Consuming Generative Models Go MAD. (2024). Masters thesis, Rice University. https://hdl.handle.net/1911/117804en_US
dc.identifier.urihttps://hdl.handle.net/1911/117804en_US
dc.language.isoengen_US
dc.rightsCopyright is held by the author, unless otherwise indicated. Permission to reuse, publish, or reproduce the work beyond the bounds of fair use or other exemptions to copyright law must be obtained from the copyright holder.en_US
dc.subjectgenerative modelsen_US
dc.subjectartificial intelligenceen_US
dc.subjectAIen_US
dc.subjectself-consumingen_US
dc.subjectautophagousen_US
dc.subjectself-trainingen_US
dc.subjectmadnessen_US
dc.subjectmodel autophagy disorderen_US
dc.subjectimage modelsen_US
dc.titleSelf-Consuming Generative Models Go MADen_US
dc.typeThesisen_US
dc.type.materialTexten_US
thesis.degree.departmentElectrical and Computer Engineeringen_US
thesis.degree.disciplineEngineeringen_US
thesis.degree.grantorRice Universityen_US
thesis.degree.levelMastersen_US
thesis.degree.nameMaster of Scienceen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
CASCO-RODRIGUEZ-DOCUMENT-2024.pdf
Size:
12.48 MB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 2 of 2
No Thumbnail Available
Name:
PROQUEST_LICENSE.txt
Size:
5.85 KB
Format:
Plain Text
Description:
Download
No Thumbnail Available
Name:
LICENSE.txt
Size:
2.99 KB
Format:
Plain Text
Description:
Download
Collections
Rice University Theses and Dissertations