Cetin: ‘We have a huge opportunity here’ in data science
April 2, 2020
Mujdat Cetin will become the next Robin and Tim Wentworth Director of the Goergen Institute for Data Science (GIDS) at the University of Rochester. The appointment is effective July 1, pending approval of the University Board of Trustees.
Cetin, an associate professor of electrical and computer engineering, has served as the interim director of GIDS since July 1, 2019. He replaces Henry Kautz, the founding director of the institute. Kautz stepped down in early 2018 to serve as director of the Division of Information and Intelligent Systems at the National Science Foundation. Prior to Cetin, Ehsan Hoque, assistant professor of computer science and the Asaro Biggar Family Fellow in Data Science, served as interim director.
"We have a huge opportunity here. We have many strengths at the University that are very well aligned with data science," Cetin says. "So, this is very exciting for me. As interim director, I have sensed how much emphasis the University places on GIDS, the importance associated with data science, and the enthusiasm for it.
"This is an evolving domain, and we face a lot of competition. Other universities are making major moves in data science. So, our challenge is: How do we position ourselves?"
Wendi Heinzelman, dean of the Hajim School of Engineering and Applied Sciences, who co-led the search for a new director, says: "Mujdat is not only an internationally-recognized researcher, but he has a wealth of leadership experience within his research field. I know he will be an outstanding leader who will continue to advance GIDS as not only a University priority but also as a national and international presence in the field of data science."
"We could not be more thrilled that Mujdat has agreed to serve as the Wentworth Director of GIDS," says Gloria Culver, dean of the School of Arts and Sciences, who also co-led the search. "Even in the very short time he has been at the University he has made a large impact on research and community -- none more so than the impact he had as interim director of GIDS. We anticipate great things for GIDS and Mujdat in the very near future."
GIDS was founded in 2014 as part of the University’s $100 million commitment to making data science its top research priority. Located in Wegmans Hall, the institute serves as a hub for interdisciplinary data science research, interdepartmental data science academic programs, and collaborations with local companies.
The institute offers educational programs ranging from summer programs for high school students to research projects engaging PhD students. It currently enrolls 130 bachelor’s and 50 master’s students whose training culminates in company-sponsored capstone projects that engage students in tackling "real-world" data science problems. In addition, the Institute recently launched a new advanced data science certificate program for professionals.
The Rochester Data Science Consortium works with more than two dozen local companies on providing timely solutions to pressing data science problems. The consortium employs 10 data and research scientists who are experts in the use of artificial intelligence, machine learning, and advanced analytics.
The New York Center of Excellence supports longer term research involving faculty and industry, commercialization of foundational data science research, and connecting students to opportunities in data science.
Heinzelman praised the "new level of activity and success" that Cetin has brought to the research program in GIDS during his time as interim director.
For example, he has established working groups among the more than 70 University faculty members affiliated with the Institute. The groups are working to develop data science research collaborations in:
- Machine learning and artificial intelligence
- Imaging, optics, and computer/human vision
- Life sciences and biomedical data science
- Healthcare analytics and digital health
- Human-data-system interfaces, including AR/VR, and robotics
- Complex systems and network data science
- Economics and business data analytics
In collaboration with Cornell University, Cetin obtained $1.5 million in National Science Foundation funding to establish a Greater Data Science Cooperative Institute. The goal is to develop theoretical foundations of data science that transcend individual fields, with a special focus on applications for health care.
He also is principal investigator of a $1.5 million NSF grant that will enable 62 doctoral students to be trained in the skills needed to advance augmented and virtual reality (AR/VR) technologies. The program will also help the students gain an appreciation for the broader cultural and societal implications of the technologies.
Cetin’s agenda as the new GIDS director is an ambitious one. In an interview he listed these top priorities:
- Expanding the opportunities for undergraduates and graduate students to become engaged in the institute’s activities, beyond their academic studies, to build and maintain an interactive, broad data science community on campus.
- Expanding and improving the educational program by exploring online offerings for professionals and traditional students, modifying and adapting courses to meet career needs of students, and improving local and global recruitment efforts.
- Hiring more data science faculty in collaboration with other departments, to teach courses and engage in research.
- Continuing the process of identifying and building on the University’s data science research strengths through core working groups. Also, generating seed money for promising projects, which, in turn can help leverage external federal, corporate, and foundation grants.
- Strengthening industry engagement in collaboration with the Consortium and Center of Excellence, "such that we relate our academic work on campus with the interests of industry through student projects as well as research proposals and grants."
- Continually reviewing the institute’s administrative structure, budget, and academic resources to ensure it can maintain and expand its competitiveness.
- Establishing partnerships with other universities, industries, foundations, and funding agencies.
"We are a great institute, but we are still a small institution," Cetin says. "So, we need to build strategic partnerships, especially focusing on areas in which we have complementary strengths." The grant with Cornell is an example, he adds.
Cetin joined the University in 2017 after serving as a faculty member at Sabanci University in Istanbul, Turkey for 12 years. At Sabanci he directed the Signal Processing and Information Systems Laboratory. From 2001 to 2005, he was with the Laboratory for Information and Decision Systems, MIT. Cetin has held visiting faculty positions at MIT, Northeastern University, and Boston University.
His research interests are within the broad area of data, signal, and imaging sciences, with cross-disciplinary links to several other areas in electrical engineering, computer science, and neuroscience. The overarching theme of his research is the development of probabilistic and machine learning-based methods for robust and efficient information extraction at various levels of abstraction from observed uncertain, complex data.
Cetin received his BS in electrical engineering from Bogazici University, Istanbul, Turkey in 1993, an MS in electrical engineering from the University of Salford, Manchester, UK in 1995, followed by a PhD in electrical engineering from Boston University, Boston, MA in 2001. He is a Fellow of the IEEE.
Data science at the University of Rochester is supported by an aggregate computational capacity that equals 420 teraflops or 420 trillion calculations per second—the equivalent computing power of more than 20,000 laptops. The Center for Integrated Research Computing (CIRC) provides computational technology and support services to more than 900 faculty members across the University. The Health Sciences Center for Computational Innovation (HSCCI) houses the health sciences research program. The VISTA Collaboratory is a 1,000-square-foot visualization lab that renders massive data sets, helping researchers visualize and analyze complex data instantly and collaboratively.