Computer Science Colloquia
Distinguished Speaker Series
Wednesday, May 13, 2015
Farinaz Koushanfar, Associate Professor, Rice University
10:45 AM, Rice Hall, Rm. 242
HOST: Dept. of Computer Science
Efficient computing and outsourcing of big and dense data analytics
Data analytics on massive and often sensitive contents regularly arise in various contemporary settings ranging from cloud computing and social networking to online services, mobile applications, and distributed processing. In this talk, I present novel computer engineering-based solutions that enable efficient and scalable explorations of the underlying patterns and dependencies present across a dense, structured datasets, with a focus on distributed and heterogeneous computing. The talk addresses the challenge of minimizing the computing, storage, and communication overhead of a broad class of iterative data analysis algorithms, down to the limits of data subspaces and the underlying heterogeneous platform. Our new methods enable optimizing for hardware acceleration as well as real-time stream processing by pushing the limits of costly data analytics to the theoretical bounds. I will show how our approach simultaneously enables efficient and secure outsourcing of computations. To facilitate automated end-to-end implementation, we provide a number of user-friendly APIs supported by our custom libraries. Proof-of-concept evaluations on several massive and dense datasets with billions of nonzero in the correlation matrix demonstrate between 1 to 2 orders of magnitude improvement in runtime, energy, and memory footprint compared with the prior art. I discuss how our findings will enable practically addressing several known classical challenges as well as exciting applications.
Bio: Farinaz Koushanfar is currently an Associate Professor with the Department of Electrical and Computer Engineering, Rice University, where she directs the Adaptive Computing and Embedded Systems (ACES) Lab. She also serves as the: principal director of the TI DSP Leadership University program; and, as the associate partner of the Intel Collaborative Research Institute for Secure Computing. She received her Ph.D. degree in Electrical Engineering and Computer Science from University of California Berkeley. Her research interests include embedded/cyber-physical systems (CPS) security, hardware trust, adaptive and customizable systems design, and secure computation. Professor Koushanfar received a number of awards and honors for her research, mentorship, and teaching including the PECASE from president Obama, ACM SIGDA Outstanding New Faculty Award, National Academy of Sciences (NAS) Kavli fellowship, Cisco IoT Security Grand Challenge award, MIT Technology Review TR-35, Young faculty/CAREER awards from NSF, DARPA, ONR, ARO, and a Best Paper Award at ACM SIGMOBILE (Mobicom).