Two faculty members from the School of Electrical and Computer Engineering at Georgia Tech have received three prestigious junior faculty awards for their work in the field of big data. Simply defined, big data involves extracting vital knowledge from massive quantities of digital information.

Christopher J. Rozell and Mark A. Davenport, both ECE assistant professors, received National Science Foundation (NSF) CAREER Awards, and Davenport also received an Air Force Young Investigator Award. They are both members of the Center for Signal and Information Processing, considered one of the world’s most preeminent groups of digital signal processing academic and research faculty.

“It is truly a major achievement for Mark and Chris to receive these three highly regarded awards in the area of big data within literally a matter of days,” said Steven W. McLaughlin, the Steve W. Chaddick School Chair of ECE. “I really look forward to watching them continue to enhance their already excellent capabilities as researchers and educators in this emerging, exciting field.”

A member of the ECE faculty since 2008, Rozell received an NSF CAREER Award for his project entitled "Exploiting Low-Dimensional Structure in Data for More Effective, Efficient, and Interactive Machine Intelligence." In this work, he is developing fundamental theory and algorithms for processing large-scale data, resulting in research that has potential benefits in any area of science and technology where data plays a fundamental role.

Rozell will develop mathematical approaches to improve the ability of machines to effectively recognize when multiple data examples arise from the same underlying phenomenon, efficiently reduce data sizes while still retaining crucial information for decision making, and optimally interact with people to learn about new data from human experts. While he will focus on developing general tools, Rozell will use computer vision tasks such as object labeling to provide accepted benchmarks for the resulting methods. To complement these technical objectives, he will also engage, recruit, and educate a diverse collection of students to STEM careers by developing curricular and outreach materials that illustrate how mathematics can be used in information systems.

A member of the ECE faculty since 2012, Davenport received the NSF CAREER Award for his project, "Learning from Coarse, Nonmetric, and Incomplete Data."  In addition, he was awarded a grant from the Air Force Young Investigator Research Program for his project, "Solving Inference and Inverse Problems Using Soft Data." Both projects center on developing mathematical models and algorithms for dealing with data that is "soft" in the sense that it consists of qualitative data like categories and comparisons rather than precise numerical values. Such data frequently arises in contexts where a key source of data consists of judgments made by people, as occurs in recommendation systems, personalized and predictive medicine, and personalized learning.

In this research, Davenport will study how low-dimensional models can be used in these contexts to help perform inference and tackle inverse problems where data, in addition to being “soft”, may also be dynamically changing and highly incomplete. While such challenges arise in many applications, a particularly important focus of Davenport’s research will be on the application of these techniques to personalized learning systems that aim to help automatically design customized assignments that are tailored to a student’s individual needs.

Faculty Bios

Christopher J. Rozell

Rozell holds the Demetrius T. Paris Junior Professorship in ECE. He is the recipient of the 2014 Georgia Tech Sigma Xi Young Faculty Award and the 2013 Georgia Tech Center for the Enhancement of Teaching and Learning/BP Junior Faculty Teaching Excellence Award. Prior to joining Tech, Rozell spent a year as a postdoctoral researcher at the Redwood Center for Theoretical Neuroscience at the University of California at Berkeley. While a Ph.D. student at Rice University, he held a Texas Instruments Distinguished Graduate Fellowship and the Nettie S. Autrey Memorial Fellowship. Rozell’s research and educational interests include biological and computational vision, theoretical and computational neuroscience, high-dimensional data analysis, distributed computing in novel architectures, and applications in imaging, remote sensing, and biotechnology.

Mark A. Davenport

Prior to his arrival at Tech, Davenport spent two years as an NSF Mathematical Sciences Postdoctoral Research Fellow in the Department of Statistics at Stanford University and as a visitor with the Laboratoire Jacques-Louis Lions at the Université Pierre et Marie Curie in Paris, France. While a Ph.D. student at Rice University, Davenport was a co-recipient of the 2007 Hershel M. Rich Invention Award for his work on the single-pixel camera and compressive sensing, and in 2011 he was awarded the Ralph Budd Award for the best Ph.D. thesis in the George R. Brown School of Engineering. His research and educational interests are in the areas of statistical signal processing, machine learning, low-dimensional signal models, and compressive sensing and low-rank matrix recovery.