Over the last several years, Mechanical Engineering Professor Levent Burak Kara has watched the number of students in his artificial intelligence and machine learning course grow steadily. In this graduate course, students learn fundamentals— probabilistic learning, pattern recognition, neural networks, clustering, regression, and search—to develop skills to solve engineering tasks.

Building on these concepts, Kara launched a new course this spring, Artificial Intelligence and Machine Learning–Project, in which students take the AI techniques they previously learned and apply them to real-world engineering problems.

“They’re tackling interesting problems within their own research field, and we’re using machine learning mostly to solve those problems,” says Kara.

Because the new course focuses on applying AI, it was structured as a project-based course, with the students divided into small groups. Every week, the groups met with Kara to discuss their progress, challenges they faced, and possible solutions.

AI is reshaping the research landscape and changing what engineers and scientists can do. Courses like this one allow students to go into the work force with an advantage.

“A lot of IT companies are interested in machine learning talents because if you understand the principles of machine learning, you can apply them to a wide range of interesting problems,” said Kara. “You can take the foundation, your basic skills, and apply them to a different application with very little overhead.”

“In the course, students formulate the problem, develop and test multiple approaches, and run rigorous validation studies to justify their claims. This allows students to get a better appreciation for the strengths and weaknesses of different algorithms, while enabling them to come up with interesting variations to the conventional algorithms,” said Kara.

A lot of IT companies are interested in machine learning talents because if you understand the principles of machine learning, you can apply them to a wide range of interesting problems.

Burak Kara, Professor, Mechanical Engineering

The student projects covered a wide range of topics. One project explored what constitutes a “good” and a “bad” photo, and how to use AI to compose a “good,” or aesthetically pleasing, photograph. Another group worked on a system that can deblur a selected part of an image.

“Our project could be subdivided into two parts. One was object detection, and the second was image deblurring,” said Ojas Joshi, who graduated this spring with his M.S. in Mechanical Engineering. “We combined those two things in an ensemble. The image deblurring algorithms deblurred the whole image—they don’t care about the detection part. So, we focused on improving object detection with image deblurring.”

Mechanical Engineering Ph.D. student Rebecca Tanzer along with her group predicted pollutant concentration at locations around Pittsburgh without monitors. The group set up a network of 50 sensors dispersed throughout the city to measure particles, specifically PM 2.5, an air pollutant that is a health concern at high concentration levels. The group gathered a large dataset consisting of data from their
sensors, land use data, and meteorological data and fed it into a random forest model and neural network, two techniques they learned in Kara’s prerequisite course. The algorithms then had outputs of predicted concentrations in a given area.

“You want to know what concentration you’re being exposed to in any given spot in Pittsburgh,” said Tanzer. “If you have a map on a website or an app, it could tell you how high the pollutant concentration is for where you are, and that might impact your decisions about where you go in the city.”

“It was good to learn about machine learning while applying it to a real problem,” said Tanzer. “It’s not just learning for the sake of learning.”