Congratulations to Jing Ma and her co-authors who earned the best research paper award at KDD '22, a premier scientific conference in knowledge discovery and data mining.Conference judges deemed their paper, Learning Causal Effects on Hypergraphs, as best among 1659 research papers submitted.
Ma is a Ph.D. candidate of computer science who works with Jundong Li, an assistant professor of electrical and computer engineering who holds joint appointments in computer science and the School of Data Science. Ma and Li co-authored the award-winning paper with a team from Microsoft Research; Ma collaborated with the Microsoft team during her summer 2021 internship, continuing their work when she returned to UVA Grounds.
Ma's research focuses on causal inference and graph mining, to which she brings extensive research experience in active learning, crowdsourcing and distributed computing.
Hypergraphs — a sub-field of mathematics and graph theory — model interactions in a network in which a hyperedge (or link) can connect any number of nodes. The team took a fresh approach to studying hypergraphs by looking at causality. They focus on the problem of individual treatment effect estimation on hypergraphs to estimate how much an intervention affects an outcome, for example the extent to which wearing a face mask affects COVID-19 infection.
Previous studies have addressed causality through statistical methods that look at a link and its paired nodes in isolation. Statistical approaches assume that the outcome on one individual should not be influenced by the treatment assignments on other individuals, or that cause-and-effect occurs only between pairs of connected individuals. Ma and her co-authors argue that these assumptions are unrealistic when hypergraphs are applied to real-world problems that involve group interactions.
“In this work, we investigate high-order interference modeling, and propose a new causality learning framework powered by hypergraph neural networks,” Ma said. “Extensive experiments on real-world hypergraphs verify the superiority of our framework over existing baselines.”
The research that Ma presented at KDD '22 advances Li's goal to better understand cause and effect in human decision-making in the era of big data, for which he earned a prestigious National Science Foundation CAREER award. Li will use his $600,000 five-year award to develop a suite of sophisticated algorithms and mathematical models, informed by human experience and intuition, to find cause-and-effect relationships in a huge amount of data. His work has the potential for broad applications in public health and medicine in addition to education.