Feb 03, 2012 - Claremont, Calif. -
Sponsored this year by Mitsubishi Electric Research Labs, the contest recognized undergraduates who demonstrate exceptional potential in the computer science field. Honorable Mentions were awarded to nominees whose work was considered exemplary.
Kevin Black ’12
Kevin Black ’12 received an Honorable Mention for his work in further developing a software program that helps biologists determine how two species may have co-evolved.
The program, “Jane 3,” applies computational techniques to analyze and reconcile the evolutionary histories of ecologically linked species—such as bees and flowers—to determine if and how they may have impacted each other’s development.
“Kevin worked on a particularly challenging biological event called ‘failure to diverge’ that is known to occur in evolution but has not been captured in existing software tools,” said Ran Libeskind-Hadas, computer science professor and Black’s project adviser. “He developed a very clever and elegant new algorithm and implemented it in Jane.”
To develop his algorithm, Black studied natural failure to diverge events—situations where a single species lives on multiple host species—to understand where and why they might occur. However, when he tried to integrate such events into the existing Jane algorithm, it increased its computational complexity and reduced the program’s speed. His persistence in tackling the problem resulted in the new algorithm and a life lesson.
“The project taught me that progress in research is not always immediate,” he said. “I was also able to see how helpful it can be to integrate knowledge from seemingly different fields. My background in computer science, mathematics and biology all ultimately came in handy working on this project.”
Stuart Pernsteiner ’12
Stuart Pernsteiner ’12 received an Honorable Mention for his work in helping to create a programming model that makes it easier to develop software that fully utilizes multi-core processors.
The project, “Observationally Cooperative Multithreading,” provides a simpler approach to parallel programming, where the computer handles many of the subtle details.
“OCM requires a lot of software support to handle all the concurrency details and Stuart worked on several different approaches to implementing the OCM model,” said Christopher Stone, associate professor of computer science, who with Melissa O’Neill, associate professor of computer science, served as Pernsteiner’s co-adviser. “He wrote thousands of lines of Haskell code and also served as a mentor for the other research students.”
Pernsteiner worked two summers on the project, which is funded by a three-year National Science Foundation grant. He created two implementations of the OCM programming model and a variety of benchmarking tools to measure their performance. The experience deepened his knowledge and influenced his post-graduation plans.
“I gained a much better understanding of computer science research and the processes involved,” he said. “This was an important factor in my decision to attend graduate school and pursue a doctorate after I graduate Mudd.”
Media Contact: Judy Augsburger