He receives grant to improve performance of deep learning models

Jingrui He
Jingrui He, Associate Professor

Associate Professor Jingrui He has been awarded a two-year, $149,921 grant from the National Science Foundation (NSF) to improve the performance of deep learning models. For her project, "Weakly Supervised Graph Neural Networks," she will focus on the lack of labeled data in Graph Neural Networks (GNNs), a deep learning method designed to perform inference on data described by graphs.

A graph is a structured way to represent data, with nodes representing entities and edges representing the relationships between these entities. GNNs provide an easy way to conduct node-level, edge-level, and graph-level prediction via machine learning. However, they usually require a large amount of label information to train the model parameters. According to He, the lack of labeled data in graphs can render many existing deep learning models ineffective in achieving the desired performance. Her new project involves a work-around so that GNNs can use unlabeled data and other relevant information.

"For example, in fraud detection, the number of known fraudulent transactions is usually very small compared to the total number of transactions, hence the lack of labeled data. Most existing GNN models tend to suffer from such label scarcity. In my new project, we aim to address this issue by leveraging weak supervision or additional information (besides the limited label information), such as labeled data from other related applications and/or access to a domain expert, in order to compensate for the lack of labeled data," said He.

In addition to fraud detection, areas such as agriculture and cancer diagnosis could also benefit from this research. He’s project will lead to a suite of new models, algorithms, and theories for constructing high-performing GNNs with weak supervision, and for understanding the benefits of weak supervision with respect to the model generalization performance and sample complexity.

He's general research theme is to design, build, and test a suite of automated and semi-automated methods to explore, understand, characterize, and predict real-world data by means of statistical machine learning. She received her PhD in machine learning from Carnegie Mellon University.

Updated on
Backto the news archive

Related News

Lee selected for leadership institute

MS/LIS student Kyra Lee had the opportunity to network with leaders in the LIS field at the 2022 Black Caucus American Library Association (BCALA) Leadership Institute. At the inaugural event, which took place from April 12-14 in Durham, North Carolina, LIS students and early career library professionals gathered for workshops, panels, facilitated discussions, and presentations. Lee was one of eighteen students selected to participate in the institute.

Kyra Lee

2021 Downs Intellectual Freedom Awards given to #FReadom Fighters and ALA Office for Intellectual Freedom staff

For libraries and librarians, 2021 was an especially challenging year in terms of the increase in attempts at censorship. According to the American Library Association (ALA) Office for Intellectual Freedom, the number of challenges to library materials more than tripled from 2020 to 2021. In addition, current estimates show that 82 to 97 percent of challenges go unreported, suggesting that the total number of challenges are significantly greater.

#FReadom Fighter logo

New project to improve health of patients with kidney failure

There are approximately 600,000 individuals in the U.S. who are undergoing hemodialysis (HD) therapy for kidney failure. In hemodialysis, a machine filters wastes, salts, and fluid from the blood when an individual's kidneys are no longer healthy enough to do this work adequately. While lifestyle changes such as getting more exercise and making better nutritional choices would benefit HD patients, they are not popular with patients—leading to poor health outcomes. A new project, led by Assistant Professor Jessie Chin, aims to boost HD patients' commitment to exercise through a long-term motivational interviewing conversational agent (LotMintBot).

Jessie Chin

iSchool alumni named 2022 Movers & Shakers

Five iSchool alumni are included in Library Journal’s 2022 class of Movers & Shakers, an annual list that recognizes 41 professionals who are moving the library field forward as a profession. Jeanie Austin (PhD '17) was honored in the Advocates category, Van McGary (MS/LIS '18) was honored in the Change Agents category, Elisandro Cabada (MS/LIS '17) and Robin Davis (MS/LIS '12) were honored in the Innovators category, and Barbara Alvarez (MS/LIS '12) was honored in the Educators category.

Spectrum Scholar Spotlight: Brandi Hart

Eight iSchool master's students were named 2021-2022 Spectrum Scholars by the American Library Association (ALA) Office for Diversity, Literacy, and Outreach Services. This "Spectrum Scholar Spotlight" series highlights the School's scholars. MS/LIS student Brandi Hart earned her BA degree in history and classical studies from Denison University.

Brandi Hart