For his senior honors project last year, titled An Information Theoretic Framework for Eukaryotic Gradient Sensing, double-degree student Joseph Kimmel ’06 developed a new way of applying mathematics and computer science to a problem in neuroscience.
Under the tutelage of Richard Salter, professor and chair of Oberlin’s computer science program, and Peter Thomas, assistant professor of mathematics, Kimmel’s project morphed into a paper co-authored by all three. The group submitted the paper for presentation at the 20th Annual Conference on Neural Information Processing Systems, a prestigious venue for mathematical neuroscientists. It was accepted.
“Work by undergraduates is rarely accepted for the conference, which rigorously reviews submissions,” Salter says. “The acceptance rate for all papers this year was only around 25 percent. The paper is also an example of interdisciplinary efforts the College so strongly encourages and values.”
Kimmel, who majored in computer science in the College and composition in the Conservatory, is pursuing graduate work in computational neuroscience this fall at the University of Chicago. His close working relationship with Salter and Thomas was invaluable. “I learned a lot about the overhead that goes along with being a research scientist: grant writing, balancing, multiple collaborations and committees, and lots of back-and-forth with funding agencies like the NSF,” he says. “Besides exploring mathematics, biology, and computer science, there was a lot of joking around and enjoying ourselves. I really feel like we became friends.”
The project, he says, put him far ahead of the curve in graduate school. “It takes most of a year to launch a research project, develop it, and progress enough to produce publishable results. Doing the honors project enabled me to accomplish something that students one and even two years ahead of me have yet to do.
“But that’s half the point of going to a school like Oberlin, isn’t it?” he adds. “Where you’re not competing with grad students for access to lab spots and faculty time? The other half is to have fun.”
Kimmel will head to Vancouver in December for the conference, where he expects to meet leading scientists and participate in tutorials in a range of subjects, including neuroscience, learning algorithms and theory, bioinformatics, image processing, and data mining.