On deep learning in physics

Machine learning, and most notably deep neural networks, have seen unprecedented success in recent years due to their ability to learn complex nonlinear mappings by ingesting large amounts of data through the process of training. This learning-by-example approach has slowly made its way into the physical sciences in recent years. In this dissertation I present a collection of contributions at the intersection of the fields of physics and deep learning. These contributions constitute some of the earlier introductions of deep learning to the physical sciences, and comprises a range of machine learning techniques, such as feed forward neural networks, generative models, and reinforcement learning. A focus will be placed on the lessons and techniques learned along the way that would influence future research projects.

Name thesis:degree_name

Doctor of Philosophy (PhD)

Discipline thesis:degree_discipline

Nuclear Engineering

Grantor

University of Ontario Institute of Technology

Year dc:date.issued

2021

Author dc:creator

• Mills, Kyle

Advisors dc:contributor.advisor

• Tamblyn, Isaac
• van Veen, Lennaert

• Machine learning
• Reinforcement learning
• Materials science
• Deep neural networks
• Deep learning

Language dc:language.iso

en

Handle dc:identifier.uri

https://hdl.handle.net/10155/1299

OAI identifier oai:identifier

oai:ontariotechu.scholaris.ca:10155/1299