Carson Easterling plans to specialize in control theory once he begins graduate school at Auburn this fall but wanted extra preparation.
Applied Statistics and Machine Learning (ELEC 5970 6970 600), an interdisciplinary course housed in the Department of Electrical and Computer Engineering, introduces students to foundational and advanced machine learning tools through hands on, application driven projects.
It gave Easterling what he needed.
“It’s invaluable to have an understanding of the foundations of artificial intelligence (AI) models and deeper knowledge of why they do certain things,” said Easterling, an electrical engineering senior who will graduate in May. “From an electrical engineering perspective, machine learning is a nice way to create models of very complex things when you have the data.”
Yin Sun, the Godbold Associate Professor in the Department of Electrical and Computer Engineering, co-directs the course with Rui Chen, the Research Extension Assistant Professor at Tuskegee University. The class, in its fourth iteration, serves 23 Auburn students and nine from Tuskegee.
“Our class covers a range of machine learning algorithms including K nearest neighbor, supporting vector machine, decision trees and neural networks, including both convolutional neural networks for computer vision and transformer neural networks that are the foundation for ChatGPT,” he said. “Simple machine learning algorithms are typically good for small dataset problems and more cutting-edge deep learning algorithms are good for big data problems. Because the students are from both engineering and agriculture backgrounds, they may face different problems.”
Sun said the course encourages both students and faculty to think more broadly about how AI can be applied.
“AI itself is interdisciplinary and all industries and private businesses in Alabama will need it,” he said. “We will continue to see the importance of AI in different fields. During these years of teaching this course, I started to work on new projects regarding AI for agriculture, AI for education, AI for 6G wireless and AI for robotics with NVIDIA and several professors at Auburn University and Tuskegee University. By teaching this course, it gives us a chance to contribute to this growing AI trend.”
Like many classes at the conclusion of a semester, Sun’s course featured poster presentations in Broun Hall before faculty and peers on April 22.
They were complex, to say the least.
What does it take to build temporally consistent real time video captioning? How can one create an interactive large language model tokenization analyzer that reveals how models break text apart? What about designing automatic video highlight detection and captioning for long form footage?
“The poster presentations make the project formal and gets the students excited,” Sun said. “Auburn engineering students are very capable. When they are serious about a course project, the project outcomes are very good. These presentations are great experiential learning activities for the students, and they are now prepared for the new AI based engineering career.”
Sam Chamoun, a teaching and research assistant who will receive his master’s degree in electrical engineering in May, pointed out the project’s value beyond Auburn.
“I am grateful for this course because it allowed me to learn machine learning while immediately applying it to a real-world project,” he said. “The collaborative structure was especially valuable. It taught me how to effectively manage and split up technical tasks for a major coding project, which is a practical skill I haven’t had the chance to develop in other courses.”
Easterling said there are few substitutes for hands on learning. He pointed to applications like robotics and simulation, where software trained models can be moved onto physical systems, and said the class helps make that transition possible by turning abstract equations into code and results.
“When you are in the classroom, you are looking at the linear algebra on the whiteboard and you are like, ‘What is going on?’ But once you get into code and you start using data and you start seeing the outputs, you start really putting the theory into practice, and that internalizes it for you. That’s where this class stands out.”
