Contact: Byron Spice
Carnegie Mellon's Project Olympus
Receives Grant From Heinz Endowments
Incubator Lab Will Boost Economic Development in Western Pennsylvania
PITTSBURGH—The Heinz Endowments has awarded $400,000 to Carnegie Mellon University's School of Computer Science to launch Project Olympus, a new initiative to chart exciting directions for the next generation of computing.
The project, directed by Distinguished Career Professor of Computer Science Lenore Blum, is designed to bring the university's researchers together with innovators at major technology companies in the Pittsburgh area, such as Seagate, Google, Microsoft, Apple and Intel.
"We expect to not only produce new science and technologies but to also move those discoveries into local companies or new spin-offs for commercial development," Blum said. "We're convinced that the ideas hatched here can also be brought to fruition here, so the talented students who graduate from our universities can pursue careers in our region."
Project Olympus will build on the success of Carnegie Mellon's five-year-old ALADDIN Center, a multidisciplinary group co-directed by Blum that applies new problem-solving methods and theoretical ideas to real-life problems. Among its innovations was CAPTCHAs, a test involving distorted letters and numbers that Web sites such as Yahoo! use to block malevolent computer programs.
The Olympus team will employ groups of students, faculty and industry researchers that focus on particular areas. The first of these "probes" will explore a technique called human computation that was pioneered by Luis von Ahn, assistant professor of computer science and a 2006 MacArthur Foundation "genius grant" winner. The technique uses multiplayer Internet games to harness human brainpower to solve problems, such as comprehension of images that remain beyond the capabilities of digital computers. Another probe will explore how to best exploit the coming generation of "multi-core" computer chips, which contain far more computer processors than the one or two processors common on today's computer chips. Plans also call for a probe on how to develop and verify control software for safety-critical medical devices, such as drug infusion pumps.
Computing will increasingly involve the accumulation, management and searching of huge stores of information, increased interconnectedness of computer users and, at the high end of performance, computers capable of quadrillions of calculations each second. These capabilities, Blum said, offer both challenges and opportunities and will require new levels of fundamental theory and innovation by computer scientists.