Electronic Thesis and Dissertation Repository

Thesis Format



Master of Science


Computer Science

Collaborative Specialization

Artificial Intelligence


Boyu Wang


The popularity of machine learning algorithms has increased in recent years as data volumes have risen, algorithms have advanced, and computational power and storage have improved. EEG-based sleep staging has become one of the most active research areas over the last decade. Labeling each sleep stage manually is a labor-intensive and time-consuming process that requires expertise, making it susceptible to human error. In the meantime, training models on an unseen dataset remains challenging due to physiological differences between subjects and electrode sensor configurations. Unsupervised domain adaptation approaches may provide a solution to this problem by borrowing knowledge from a labeled dataset to train an unlabeled dataset. A source-free unsupervised domain adaptation methodology is employed in this thesis to solve the problem of automatic single-channel EEG sleep stage classification. Our study shows that pre-training source domain models followed by supervised fine-tuning improve the learned representations when applied to EEG sleep signals. We further develop weighted diversity loss in order to achieve a model that outperforms state-of-the-art unsupervised domain adaptation techniques without access to source domain data.

Summary for Lay Audience

The classification of sleep stages refers to the process of identifying the different stages of sleep that an individual experiences during the course of the night. Sleep has five different stages, each characterized by distinct patterns of brain activity, muscle tone, and other physiological indicators. It is essential to classify the stages of sleep in order to diagnose sleep disorders and related diseases. In spite of this, categorizing sleep stages manually is inefficient in terms of time and requires expert assistance. As a result, there has been extensive research concerning the use of machine learning to classify sleep stages automatically. This approach can be helpful in a variety of applications, such as developing sleep-tracking devices and applications. However, due to individual variability and the configurations of the sensors that collect sleep data, automated sleep stage classification is not always accurate. This is why researchers have employed domain adaptation techniques in automated sleep stage classification. It is a technique of leveraging knowledge from a related domain (also known as the source domain) to enhance the performance of a model in another domain (also known as the target domain). This thesis directly addresses this shortcoming by developing a novel approach to training a model to perform sleep stage classification automatically. It will investigate the factors that contribute to developing a methodology that uses the source domain knowledge to classify unlabeled target data while maintaining the privacy of its subjects.