Skip to main content

Department Receives NSF Career Awards

Rui Song and Eric Laber, associate professors in the Department of Statistics, recently received National Science Foundation (NSF) CAREER Awards for their research. NSF awards these prestigious CAREER grants to outstanding junior faculty members to help them advance their research and teaching activities.

The Department is proud to have two recipients of the CAREER award in the same year, which is unprecedented. We acknowledge the amazing work that Marie Davidian, our director for research, in collaboration with Ann Zhang, is doing in helping our faculty become incredibly successful with their grant-writing initiatives.

Song received the award for her for work entitled “Semiparametric and Machine Learning Approaches for Big Data Challenges in Precision Medicine.” The main objective of her proposal is to develop cutting-edge and powerful semiparametric methods and machine learning tools to realize the promise of precision medicine via developing optimal individualized treatment regimes (ITR). Specifically, she aims to develop flexible and efficient and methods for discovering optimal ITRs, develop a general class of optimal ITRs, develop optimal ITRs with high-dimensional data, and develop optimal ITRs under population heterogeneity. She plans to integrate the research goals with her educational activities by supervising and training students, developing new courses, revising existing courses and providing outreach to K-12 educators and the scientific community and industry through collaborations.

“The proposed work will contribute to both fields of semiparametric inference and machine learning. Machine learning methods have rarely been studied for doubly-robust estimation and optimal ITRs with NP-dimensionality,” Song said. “The theoretical developments that include driving nonasymptotic distribution, risk bounds and new empirical process technical tools are challenging. The theory to be developed in this project will be fundamentally important and generally applicable for studying semiparametric models in the high-dimensional setting.”

Using semiparametric and machine learning methods for precision medicine is an emerging novel area. “The clinical findings from my collaboration with medical practitioners will lead to major progress in addressing important clinical questions on the treatment recommendations for breast cancer patients, colon cancer patients and colorectal cancer patients,” Song said. Her proposed work is directly related to the White House Precision Medicine Initiative and she expects that it will help accelerate the discovery of new personalized treatment strategies.

Laber received a CAREER award for his work entitled, “Big computation and the management of emerging infectious diseases.” The methodology proposed in the grant can be used to translate real-time data on emerging infectious diseases, like the Middle Eastern Respiratory Syndrome and superbugs, into recommendations about where, when and to whom to apply interventions so as to minimize the negative impacts of the disease while reducing overall resource consumption. The grant has three primary subprojects:

  1. online, simulation-based planning,
  2. interpretable allocation strategies using priority lists
  3. approximation error bounds that account for computational expenditure.

“A common thread through these three subprojects is the connection of difficult statistical estimation problems with well-established algorithms in machine learning,” Laber said.

Outreach to the community is a major educational objective. “We are using artificial intelligence in videogames to teach high-school students and undergraduates about real-time data-driven decision making. My students Maria Jahja, Nick Meyer, Robert Pehlman, Shuping Ruan and Longshaokan Wang have been doing amazing work on developing games and tutorials,” Laber said.

His long-term goal is to create a principled framework for translating heterogeneous data streams on the current status of an epidemic into recommendations about optimal treatment allocation. “Such a framework could have tremendous positive impacts on public health across the world,” he said.