Deep learning in health-care with mixture model : a simulation study
Through the research and learning in this Final Year Project (FYP), an innovative deep mixture model: Deep Gaussian Mixture Model (DGMM), for clustering problem is being fully studied and understood by a theoretical derivation and a reusable, efficient and code-optimized implementation of the Stocha...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2018
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/76215 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Through the research and learning in this Final Year Project (FYP), an innovative deep mixture model: Deep Gaussian Mixture Model (DGMM), for clustering problem is being fully studied and understood by a theoretical derivation and a reusable, efficient and code-optimized implementation of the Stochastic Expectation Maximization (SEM) algorithm on DGMM in Python. On top of that, an extension of the current DGMM model is also being explored and implemented which enables the model to fit high dimensional data through a proposed model selection methodology which uses high dimensional Principal Component Analysis (PCA) and Principal Orthogonal Complement Thresholding (POET) to select the optimal hyper-parameters and achieved a promising result. |
|---|