Our congratulations go to three CSE faculty who have been selected for NSF CAREER Awards in 2016: Jason Mars for his work in architectures for intelligent assistants, Barzan Mozafari for his work in designing predictable databases, and Jenna Wiens for her use of machine learning in critical care.
Prof. Jason Mars was awarded an NSF CAREER grant for his project, “CAREER: Advancing the Frontier in System Architectures for Artificially Intelligent Services and Applications.”
The award will enable Prof. Mars to understand how future cloud and mobile systems should be designed to support increasing demand from users of intelligent assistants. Intelligent assistants require sophisticated machine learning and computer visions algorithms and he proposes to make the current computing platforms more efficient and expand systems to make them more intelligent, while also allowing for future research.
His work will impact national interests, economic advancement, technology, as well as, innovation in undergraduate and graduate education.
More information about the project is available in Prof. Mars’s CAREER Award Posting by NSF.
Prof. Mars' current research interests include cross-layer systems in software and hardware, datacenter and warehouse-scale computer architecture, and hardware/software co-design.
Prof. Mars received his Ph.D in Computer Science at The University of Virginia in 2012 and joined the faculty at Michigan in 2013. Before joining U-M, he was an assistant professor in the CSE department at The University of California, San Diego. Also, he has served as visiting scientist at Google, which involved investigating opportunities to improve efficiency of Google’s backend infrastructure.
He has also received numerous honors and awards including the Preuss Faculty Scholar Appointment, the UVA Research Award, and Best Paper Awards from CGO ’12 and Computer Architecture Letters.
Prof. Barzan Mozafari was awarded an NSF CAREER grant for his research project, "CAREER: Designing a Predictable Database - An Overlooked Virtue."
Four decades of research on database systems has focused primarily on the improvement of average raw performance, with little emphasis placed on the predictability of database management systems. However, as database systems have become more complex, their erratic and unpredictable performance has become a major challenge facing database users and administrators alike. With the increasing reliance of mission-critical business applications on their databases, maintaining high levels of database performance (i.e., service level guarantees) is now more important than ever. Cloud users find it challenging to provision and tune their database instances, due to the highly non-linear and unpredictable nature of today’s databases. Even for deployed databases, performance tuning has become somewhat of a black art, rendering qualified database administrators a scare resource.
Under this project, Prof. Mozafari aims to restore the missing virtue of predictability in the design of database systems. First, he will quantify the major sources of uncertainty in a database in a principled manner. Then, by rethinking the traditional design of a database system, he will architect a new generation of databases that treat predictability as a first class citizen in their various stages of query processing, from physical design to memory management and query scheduling. Moreover, to accommodate existing database systems (which are not predictable by design), he will develop tools and methodologies for predicting their performance more accurately.
Commenting on the project, Prof. Mozafari says that "Through the process of building a predictable database in a bottom-up fashion and in a principled manner, we can expect great insight into improving existing database products and can instigate a radical shift in the way that future databases are designed and implemented."
Prof. Mozafari is passionate about building large-scale data-intensive systems that are more scalable, more robust, and more predictable, with a particular interest in database-as-a-service clouds, distributed systems, and crowdsourcing. In his research, he draws on advanced mathematical models to deliver practical database solutions, adapting concepts and tools from applied statistics, complexity theory, automata theory, and machine learning.
Prof. Mozafari received his PhD in Computer Science from the University of California Los Angeles in 2011. He joined the faculty of CSE at the University of Michigan in 2013 after two years as a postdoctoral researcher at Massachusetts Institute of Technology in the CSAIL Lab. Prof. Mozafari has won several awards and fellowships, including Best Paper Awards at SIGMOD 2012 and EuroSys 2013. He is affiliated with the Database Research Group and the Software Systems Lab in CSE and the Center for Data-Driven Computational Physics at the Michigan Institute for Computational Discovery and Engineering (MICDE).
Prof. Jenna Wiens was awarded an NSF CAREER grant for her research project, "CAREER: Adaptable, Intelligible, and Actionable Models: Increasing the Utility of Machine Learning in Clinical Care."
In recent years, the availability of clinically relevant medical datasets has grown enormously. However, there have been relatively few successes regarding translation to practice, and clinicians still base the bulk of their daily decisions on relatively small amounts of patient-specific data.
This research aims to transform the larger realm of available data into actionable knowledge through the exploration of new fundamental research directions and approaches in machine learning. By targeting patients identified as high-risk through computational data-driven models, practitioners could reduce the burden of disease in a cost-effective manner.
For data-driven predictive models to become widely and safely adopted in clinical care, there remain several key research challenges that the machine learning community must address: poor adaptability to complex unexpected changes in patient populations and clinical protocols, insufficient intelligibility of accurate but uninterpretable models, and absence of actionability, with accuracy overcoming actionability. Prof. Wiens will address these and other issues under the umbrella of this project.
More information about the project is available in Prof. Wiens' CAREER Award Posting by NSF.
Prof. Wiens' primary research interests lie at the intersection of machine learning and healthcare. The overarching goal of her research agenda is to develop the computational methods needed to help organize, process, and transform data into actionable knowledge. In addition to her work in healthcare, she develops machine learning methods for the extraction of strategically useful information from player tracking data in the National Basketball Association.
Prof. Wiens received her PhD in Electrical Engineering and Computer Science from the Massachusetts Institute of Technology in 2014 and joined the faculty of CSE at the University of Michigan that year. Her PhD research focused on developing accurate patient risk-stratification approaches that leverage spatiotemporal patient data, with the ultimate goal of discovering information that can be used to reduce the incidence of healthcare-associated infections. She is affiliated with the Artificial Intelligence Laboratory at Michigan.
About the NSF CAREER Award
The CAREER grant is one of the National Science Foundation's most prestigious awards, conferred for "the early career-development activities of those teacher-scholars who most effectively integrate research and education within the context of the mission of their organization."
Posted: February 22, 2016