In a machine-learning AI model, “you have your input and you have your output,” said Wall, who is earning his Ph.D. in data science and engineering through USD’s Department of Computer Science. “I look at what comes in-between—the little black box between the input and output.”

The field of explainable AI develops methods to open that black box makes AI more understandable and transparent to humans. Wall already has coauthored a book on this topic with KC Santosh, Ph.D. computer science professor. This July, he gave a remote presentation of his work, “Explainable AI: From Taxonomies to Modern Practice,” at the 2025 IEEE Signal Processing Society School on Explainable AI and Applications to Biometric Signal Processing at the Indian Institute of Information Technology, Allahabad, India.

Wall, who is originally from Akron, Iowa, earned his master’s degree in computer science at USD in 2022. Wall returned to USD after working as a research engineer for a year and a half performing document image analysis in France.

The desire to pursue his Ph.D. and work with Santosh brought Wall back to USD to expand on his previous work on explainable artificial intelligence and tackle other projects.

One goal of the field of explainable artificial intelligence is to establish AI’s trustworthiness.

“Especially in high-stakes domains, like the medical field, we need to build systems that people can have a certain amount of trust in,” Wall said.

In machine learning, the type of AI that enables computers to learn from large data sets without being specifically programmed, data engineers can’t “read the machine’s brain” but they can interpret what it has taught itself by analyzing patterns in its structure and outputs.

In addition to other methods, data engineers use computer programs post-hoc, or after the AI model has produced an output, to determine why an AI model made a single decision and how each feature in a dataset contributes to a prediction.

Developers can also build explainability features into the AI models by integrating principles like fairness, transparency and accountability throughout the entire development process. Some of Wall’s work falls into this category.

In USD’s research-intensive doctoral data science and engineering program, Wall has numerous projects in the works. One of his recent studies uses AI to evaluate the complex vocalizations of whales. Another focuses on pulling more information from AI models that analyze images to give better information on the model’s decision making.

Keeping up with fast pace of new developments in the field of AI keeps Wall busy while he designs and builds AI systems.

“In this field, there are new innovations all the time,” he said.

At USD’s AI Research Lab, Wall collaborates with other students and computer science faculty.

“The lab is fantastic,” he said. “It’s a perfect environment for anyone who wants to do research. I can knock on the door of any professor’s office and ask them for help. It’s a super supportive system.”

Wall is a big booster of the USD’s artificial intelligence programs and encourages students to take advantage of AI and data science and engineering offerings in the USD Department of Computer Science.

“We want to build up this area to be an AI powerhouse,” Wall said.

Press Contact
Alissa Matt
Contact Email [email protected]
Contact Website website

You May Also Like