Photo by Tony RinaldoPhoto by Tony Rinaldo
Tufts University
Computer Science
Impact of Computational Geometry on Depth-Based Statistics

Diane Souvaine is a professor of computer science at Tufts University. She works in computational geometry, a field concerned with the design and analysis of algorithms for solving geometric problems. Applications can be found in such fields as VLSI (very-large-scale integration) design, computer graphics, robotics, computer-aided design, pattern recognition, and statistics.

While at the Institute, Souvaine will address underlying computational geometry issues in the development of efficient practical algorithms for various flavors of data depth and associated metrics: open computational problems related to halfspace depth, simplicial depth, and regression depth contours; the applicability of depth contours to quantification of multivariate features of a data set and its visualization; the extension of known two-dimensional algorithms to higher dimensions; and new approaches for outlier detection and for ascertaining their validity.

Souvaine earned an AB at Radcliffe College, an AM at Dartmouth College, and MSE, MA, and PhD degrees from Princeton University. A faculty member at Rutgers University from 1986 until 1998 when she moved to Tufts University, Souvaine served from 1992 to 1994 as acting director of the Center for Discrete Mathematics and Theoretical Computer Science (DIMACS). From 2002 until 2005, she served as chair of computer science at Tufts. Souvaine conducts research in the area of computational geometry and leads activities to encourage students of diverse backgrounds to pursue technical fields.

This information is accurate as of the fellowship year indicated for each fellow.
Photo by Tony Rinaldo