Files in this item

FilesDescriptionFormat

application/pdf

application/pdfKIM-DISSERTATION-2018.pdf (4MB)
(no description provided)PDF

Description

Title:Machine learning approaches to star-galaxy classification
Author(s):Kim, Junhyung
Director of Research:Brunner, Robert J
Doctoral Committee Chair(s):Thaler, Jon J
Doctoral Committee Member(s):Gollin, George D; Schwing, Alexander G
Department / Program:Physics
Discipline:Physics
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:Ph.D.
Genre:Dissertation
Subject(s):data analysis
image processing
photometric surveys
star-galaxy classification
cosmology
deep learning
convolutional neural networks
generative adversarial networks
Abstract:Accurate star-galaxy classification has many important applications in modern precision cosmology. However, a vast number of faint sources that are detected in the current and next-generation ground-based surveys may be challenged by poor star-galaxy classification. Thus, we explore a variety of machine learning approaches to improve star-galaxy classification in ground-based photometric surveys. In Chapter 2, we present a meta-classification framework that combines existing star-galaxy classifiers, and demonstrate that our Bayesian combination technique improves the overall performance over any individual classification method. In Chapter 3, we show that a deep learning algorithm called convolutional neural networks is able to produce accurate and well-calibrated classifications by learning directly from the pixel values of photometric images. In Chapter 4, we study another deep learning technique called generative adversarial networks in a semi-supervised setting, and demonstrate that our semi-supervised method produces competitive classifications using only a small amount of labeled examples.
Issue Date:2018-01-31
Type:Thesis
URI:http://hdl.handle.net/2142/100895
Rights Information:Copyright 2018 Junhyung Kim
Date Available in IDEALS:2018-09-04
Date Deposited:2018-05


This item appears in the following Collection(s)

Item Statistics