Jun 12, 2013 - Claremont, Calif. -
Sponsored this year by Microsoft Research, the contest recognized undergraduates who demonstrate exceptional potential in the computer science field.
Jack Ma ’14
Ma received a Runner-Up designation for his work on a storage modeling and analysis project.
He investigated “deduplication,” a technique in which a storage system recognizes duplicate data saved on a computer—such as copied files or files that have been downloaded more than once—and then removes the duplicates so that only one copy is stored while the illusion of multiple copies in separate locations is maintained.
“Jack applied his algorithmic talents to the problem of analyzing the long-term behavior of typical files. This analysis is slow, so he developed MapReduce-style techniques to speed the calculations,” said Geoff Kuenning, computer science professor and Ma’s project advisor. “Jack has been a consistently innovative and productive researcher, and I believe that he shows exceptional promise for the future.”
Alexandra “Xanda” Schofield ’13
Schofield received a Finalist designation for her work implementing a new, auto-parsing algorithm for the Impro-Visor software project to help jazz musicians understand and learn new chord progressions.
Short for “Improvisation Advisor,” Impro-Visor is a music notation program designed to help jazz musicians compose and hear solos similar to ones that might be improvised.
“Impro-Visor is a fantastic piece of software trying to help jazz musicians in ways that result in a cool tool with a lot of elaborate code. It's critical to be able to navigate around that to add something new without hurting any prior pieces of it,” said Schofield. “The project forced me to be both a good theoretician and a good programmer.”
Schofield was part of a student team that worked with computer science Professor Bob Keller to develop a roadmap capability for Impro-Visor, where chord progressions are organized as “bricks,” and tunes are represented as a series of bricks.
“There are fewer types of bricks than there are tunes. Therefore, by learning to play based on roadmaps, it is easier to transfer knowledge from one tune to another,” Keller said. “Xanda contributed an idea for eliminating ambiguity based on assigning ‘costs’ to brick types and applying the costs in a minimal-cost, path-finding algorithm.”
Media Contact: Judy Augsburger