Date of Award
2021
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Complex Systems and Data Science
First Advisor
Safwan Wshah
Second Advisor
Jianing Li
Abstract
Machine learning, and the sub-field of deep learning in particular, has experienced an explosion in research interest and practical applications over the past few decades. Deep learning approaches seem to have become the preferred approach in many domains, outpacing the use of more traditional machine learning methods. This transitionhas also coincided with a shift away from feature engineering based on domain knowledge. Instead, the common deep learning philosophy is to learn relevant features through the combination of expressive models and large datasets. Some have interpreted this paradigm shift as the death of domain knowledge. I argue that domain knowledge is still broadly used in deep learning systems, and even critically important, but where and how domain knowledge is used has evolved. To support this argument I present three recent deep learning applications in disparate domains that each heavily rely on domain knowledge. Based on these three applications I discuss strategies for where and how domain knowledge is being effectively incorporated into newer deep learning systems.
Language
en
Number of Pages
196 p.
Recommended Citation
Van Oort, Colin, "Leveraging Domain Knowledge in Deep Learning Systems" (2021). Graduate College Dissertations and Theses. 1468.
https://scholarworks.uvm.edu/graddis/1468