Unsupervised Learning for Discovery in 2D & 3D Scenes: Towards Unbiased Understanding of Biomedical Images
Abstract/Contents
- Abstract
- While in recent years artificial intelligence has made impressive strides in many fields, the bulk of this progress has been limited to supervised methods and tasks with an abundance of labeled data. However, in many real-world domains, labeling may be burdensome or near impossible, and on the extreme end, annotations of interest may lie beyond the extent of human knowledge. Unsupervised learning, which broadly involves training machine learning models to encode information from unlabeled data, is then particularly promising as a means to address these problems. It has the potential to discover new objects and structures in 2D & 3D scenes beyond what humans can recognize. Such promise is especially exciting for discovery in the biomedical domain. Biological data can be vastly complex, and as new technologies help us collect more complicated data, there is increasing opportunity for unsupervised learning to help scientists turn observations into insights, shedding new perspectives on the underlying unknowns behind human health. In my thesis, I both establish unsupervised learning methods that expand the state-of-the-art in deep learning as well as apply my methods on real-world, cutting-edge biological data.
Description
| Type of resource | text |
|---|---|
| Date created | May 4, 2021 |
Creators/Contributors
| Author | Hsu, Joy |
|---|---|
| Primary advisor | Yeung, Serena |
| Advisor | Chiu, Wah |
| Degree granting institution | Stanford University, Department of Computer Science |
Subjects
| Subject | machine learning |
|---|---|
| Subject | deep learning |
| Subject | unsupervised learning |
| Subject | differential geometry |
| Genre | Thesis |
Bibliographic information
Access conditions
- Use and reproduction
- User agrees that, where applicable, content will not be used to identify or to otherwise infringe the privacy or confidentiality rights of individuals. Content distributed via the Stanford Digital Repository may be subject to additional license and use restrictions applied by the depositor.
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Preferred citation
- Preferred Citation
- Hsu, Joy. (2021). Unsupervised Learning for Discovery in 2D & 3D Scenes: Towards Unbiased Understanding of Biomedical Images. Stanford Digital Repository. Available at: https://purl.stanford.edu/nv775dt3762
Collection
Undergraduate Theses, School of Engineering
View other items in this collection in SearchWorksContact information
- Contact
- engreference@stanford.edu
Also listed in
Loading usage metrics...