Wavelet Flow: Fast Training of High Resolution Normalizing Flows

Simple item page
dc.contributor.advisorBrubaker, Marcus
dc.contributor.authorYu, Jason Jiasheng
dc.date.accessioned2020-11-13T13:49:16Z
dc.date.available2020-11-13T13:49:16Z
dc.date.copyright2020-07
dc.date.issued2020-11-13
dc.date.updated2020-11-13T13:49:15Z
dc.degree.disciplineComputer Science
dc.degree.levelMaster's
dc.degree.nameMSc - Master of Science
dc.description.abstractNormalizing flows are a class of probabilistic generative models which allow for fast density computation, efficient sampling, and are effective at modelling complex distributions like images. A drawback among current methods is their significant training cost, sometimes requiring months of GPU training time to achieve state-of-the-art results. This thesis introduces Wavelet Flow, a multi-scale, normalizing flow architecture based on wavelets. A Wavelet Flow has an explicit representation of signal scale that inherently includes models of lower resolution signals and conditional generation of higher resolution signals, i.e., super resolution. A major advantage of Wavelet Flow is the ability to construct generative models for high resolution data (e.g., 1024 1024 images) that are impractical with previous models. Furthermore, Wavelet Flow is competitive with previous normalizing flows in terms of bits per dimension on standard (low resolution) benchmarks while being up to 15 faster to train.
dc.identifier.urihttp://hdl.handle.net/10315/37900
dc.languageen
dc.rightsAuthor owns copyright, except where explicitly noted. Please contact the author directly with licensing requests.
dc.subjectArtificial intelligence
dc.subject.keywordsComputer
dc.subject.keywordsVision
dc.subject.keywordsComputer vision
dc.subject.keywordsArtificial intelligence
dc.subject.keywordsWavelets
dc.subject.keywordsHaar
dc.subject.keywordsNormalizing flow
dc.subject.keywordsInvertible neural networks
dc.subject.keywordsNeural networks
dc.subject.keywordsDeep learning
dc.subject.keywordsGenerative modelling
dc.subject.keywordsDensity estimation
dc.subject.keywordsMulti-scale
dc.subject.keywordsImage generation
dc.titleWavelet Flow: Fast Training of High Resolution Normalizing Flows
dc.typeElectronic Thesis or Dissertation

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Yu_Jason_J_2020_Masters.pdf
Size:
112.03 MB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 2 of 2
No Thumbnail Available
Name:
license.txt
Size:
1.83 KB
Format:
Plain Text
Description:
Download
No Thumbnail Available
Name:
YorkU_ETDlicense.txt
Size:
3.36 KB
Format:
Plain Text
Description:
Download

Collections

Computer Science and Engineering

All items in the YorkSpace institutional repository are protected by copyright, with all rights reserved except where explicitly noted.