IT Foundations | UDaily

Guangmo (Amo) Tong, an assistant professor in UD's College of Engineering, uses basic computer science research to improve data-driven decision-making.  With a better understanding of the theories that underpin algorithmic decision-making, Tong's research may support future breakthroughs in a wide range of applications.

Guangmo (Amo) Tong, an assistant professor in UD’s College of Engineering, uses basic computer science research to improve data-driven decision-making. With a better understanding of the theories that underpin algorithmic decision-making, Tong’s research may support future breakthroughs in a wide range of applications.

Photo by Kathy F. Atkinson | Illustrations by Joy Smoker

UD Engineering’s Guangmo (Amo) Tong Receives NSF CAREER Award for Research in Computer Science

From self-driving cars to facial recognition software, AI-powered technologies rely on powerful algorithms that can make sense of complex, real-world data. These algorithms are part of a larger computing workflow, known as a pipeline, which involves collecting data, identifying patterns, and making decisions about what to do next.

But decision-making pipelines specialized for one type of task – to recognize faces, for example – cannot be used to analyze other types of information, such as helping a smart car decide whether to turn right or left at the next intersection. . Given these limitations, how can researchers create new decision-making algorithms and pipelines that can work on a wider variety of data sets and real-world scenarios?

Tong Guangmo (Amo)assistant professor at Department of Computing and Information Science at the University of Delaware College of Engineering, uses basic computer science research to improve data-driven decision making. With a better understanding of the theories that underpin algorithmic decision-making, Tong’s research may support future breakthroughs in a wide range of applications.

Research in Tong Computational Data Science Lab focuses on two areas of computer science: the design of algorithms, the step-by-step instructions used by computer programs to accomplish specific tasks, and the development of statistical learning techniques, the types of mathematical models used to analyze and interpret large data sets.

This image shows an example of the decision pipelines that will be developed under Tong's CAREER award.

This image shows an example of the decision pipelines that will be developed under Tong’s CAREER award.

Beyond focusing on the application of algorithms and statistical models, Tong’s group is focused on gaining a deeper understanding of how algorithms make decisions using data. “We need a deeper understanding of decision-making, and this foundation can in turn improve state-of-the-art solutions for specific applications,” Tong said. “We believe that the foundation is important in that it substantiates the performance of the model by providing provable guarantees.”

Tong’s expertise in computational data science was recently recognized by the National Science Foundation (NSF) with a Faculty Early Career Development Program (CAREER) award, a prestigious grant that supports early-career faculty who have the potential to become leaders in research and education. Tong’s CAREER award includes funding of $513,468 over five years and began in September 2022.

“Different from many researchers in the community who focus on how to apply existing AI techniques to different scenarios, Dr. Tong aims to address a fundamental question of how to design and implement general algorithms of decision making that can potentially work on a variety of data sets. ,” said Weisong Shi, professor, director of the CIS department and collaborator of Tong for this award. “This task is extremely difficult but will have a much wider impact.”

Tong’s project will focus on developing decision-making pipelines for discrete decisions, or those associated with a specific action (like “turn right” or “stop”). This research will focus on both the design of algorithms and the exploration of the most appropriate statistical models given the type of data analyzed.

“At first, we plan to focus more on the theoretical parts of this research – the design of algorithms to solve optimization problems and learning pipelines,” Tong said. “After that, we plan to focus on several applications, including social network analytics and autonomous systems.”

In the area of ​​social network analysis, Tong and his group will use the award to further their work on “social contagion management,” or how different messages move through a network. By determining how information flows through social networks, Tong’s research could help researchers understand the factors that influence the dynamics of messages, behaviors and opinions.

Tong’s group will also use the award to help Gonzalo ArceProfessor Charles Black Evans at Department of Electrical and Computer Engineeringaddressing challenges related to inverse problems in complex networks, computational imaging and 3D Lidar sensing in his lab.

Here is an example of a social network (left) and an information cascade broadcast (right).  Tong and his team will study both of these structures as part of their work on “social contagion management,” or how different messages move through a network.

Here is an example of a social network (left) and an information cascade broadcast (right). Tong and his team will study both of these structures as part of their work on “social contagion management,” or how different messages move through a network.

“One of the things we’re working on is the development of hypergraphic neural networks, and what’s important is having a theoretical model of the power and expression of those networks,” Arce said. . “Our group is developing network architectures and the algorithms that drive them, but it’s nice that his group is analyzing how much more powerful these new networks are compared to the current state of the art.”

Researchers at Tong’s Computational Data Science Lab will also continue to test their newly developed algorithms on self-driving platforms such as HydraOne, an autonomous vehicle platform designed by Shi’s lab, to see how well their algorithms perform in the real world. Through this award, Tong plans to engage undergraduate and graduate students in this and other related projects that move his group’s research findings from theoretical concepts to applied settings.

The award will also support Tong’s ongoing outreach activities, with ongoing mentorship for high school students in conjunction with UD K-12 Engineering Awareness Programand will also support the development of graduate programs in computer science.

Tong said teaching graduate students the fundamentals of algorithm design and statistical modeling is key to training the next generation of data scientists and software engineers, who will work with datasets larger than that. larger ones that require new algorithms, statistical learning models, and computational approaches to make sense. complex and real data.

“It is often possible to implement algorithms in applications without understanding why it works, but I hope to be able to leverage these projects to help students focus more on the fundamentals of machine learning and algorithms by putting focus on program design,” Tong said.

Arce added that a strong foundation in theory also opens up new areas of collaboration at UD. “In machine learning, it’s important to have a well-rounded portfolio, and having the underlying applications, theory, and architectures is the ideal package,” he said.

“Dr. Tong’s recognition is the second NSF CAREER award the CIS department has received this year, after Dr. Lena Mashayekhy received her award in April,” Shi said. “Combined, nearly 40 percent of our full and departmental faculty have received this prestigious award, demonstrating the high quality of cutting-edge research by our faculty and students at the University of Delaware.”

Comments are closed.