Probability distribution construction via deep learning
The pursuit of estimating probability distributions of complex data is an ongoing challenge. Existing traditional methods impose a ceiling to the true resemblance of the targeted data distribution, due to their assumptions on the shape of the targeted data distribution. Recently, generative models h...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Year Project / Dissertation / Thesis |
| Published: |
2023
|
| Subjects: | |
| Online Access: | http://eprints.utar.edu.my/6152/1/HANNAH_TAN_E%2DLING%2D2005143.pdf http://eprints.utar.edu.my/6152/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The pursuit of estimating probability distributions of complex data is an ongoing challenge. Existing traditional methods impose a ceiling to the true resemblance of the targeted data distribution, due to their assumptions on the shape of the targeted data distribution. Recently, generative models have garnered substantial attention for its ability to replicate high-resolution images, thereby learning the distribution of high-complexity data. Inspired by this paradigmatic approach to learn a distribution without relying on an assumption about the shape of the target data distribution, this project explores the bridging of Deep Learning and Statistics within the area of distribution generation methods. This paper provides the overall context of the research problem in Chapter 1, elaborates on existing literature and related works in Chapter 2, discusses the methodology and execution plan of this project in Chapter 3, mentions the results from what was executed in Chapter 4 and lastly concludes in Chapter 5. |
|---|