Please use this identifier to cite or link to this item:
https://hdl.handle.net/10356/171953
Title: | Unsupervised learning with diffusion models | Authors: | Wang, Jiankun | Keywords: | Engineering::Computer science and engineering | Issue Date: | 2023 | Publisher: | Nanyang Technological University | Source: | Wang, J. (2023). Unsupervised learning with diffusion models. Master's thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/171953 | Abstract: | In computer vision, a key goal is to obtain visual representations that faithfully capture the underlying structure and semantics of the data, encompassing object identities, positions, textures, and lighting conditions. However, existing methods for un-/self-supervised learning (SSL) are restricted to untangling basic augmentation attributes such as rotation and color modification, which constrains their capacity to efficiently modularize the underlying semantics. In the thesis, we propose DiffSiam, a novel SSL framework that incorporates a disentangled representation learning algorithm based on diffusion models. By introducing additional Gaussian noises during the diffusion forward process, DiffSiam collapses samples with similar attributes, intensifying the attribute loss. To compensate, we learn an expanding set of modular features over time, adhering to the reconstruction of the Diffusion Model. This training dynamics biases the learned features towards disentangling diverse semantics, from fine-grained to coarse-grained attributes. Experimental results demonstrate the superior performance of DiffSiam on various classification benchmarks and generative tasks, validating its effectiveness in generating disentangled representations. | URI: | https://hdl.handle.net/10356/171953 | DOI: | 10.32657/10356/171953 | Schools: | School of Computer Science and Engineering | Rights: | This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). | Fulltext Permission: | open | Fulltext Availability: | With Fulltext |
Appears in Collections: | SCSE Theses |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Final Thesis.pdf | 1.48 MB | Adobe PDF | View/Open |
Page view(s)
404
Updated on Nov 4, 2024
Download(s) 20
329
Updated on Nov 4, 2024
Google ScholarTM
Check
Altmetric
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.