A team of one undergraduate, two mathematics graduate students, one continuing education student and two faculty members from the Department of Mathematical Sciences at the University of Delaware scored a perfect 100 in Prof. Nick Trefethen's "100-Dollar, 100-Digit Challenge." The challenge first appeared in the January/February 2002 issue of SIAM News. The Challenge consisted of ten questions for which there was no known way to express the solution in terms of elementary quantities. There were ten questions, and the challenge was to find the first ten digits for each question. Prof. Trefethen's original posting stated that he would be impressed with any team that found 50 correct digits. A copy of the original article is posted on the SIAM web site.

http://www.siam.org/siamnews/01-02/challenge.pdf

Prof. Trefethen's web site has a list of the other teams earning perfect scores as well as the five second place winners who score 99's.

http://web.comlab.ox.ac.uk/oucl/work/nick.trefethen/hundred.html

At the conclusion of the contest, twenty teams submitted perfect scores. The University of Delaware team was recognized as one of three to receive the $100 reward because the solution was the product of a unique collaboration between undergraduates, graduates and faculty.

At the time of the original posting, Profs. Toby Driscoll and Lou Rossi were teaching graduate and undergraduate numerical analysis courses and sought to spark more interest in the topic by answering the challenge. Soon, other students in the department saw the problems and were drawn into the group. Thus, continuing education student Jonathan Leighton, undergraduate Eli Faulkner, and graduate students Carl DeVore, and Sven Reichard joined the core group. Interestingly, the only numerical analysts on the team were faculty advisors Driscoll and Rossi. Eli Faulkner is interested in topology. Graduate students DeVore and Reichard are candidates in the discrete mathematics group. Jon Leighton has a variety of interests in applied mathematics and solid mechanics.

The team quickly found that direct numerical attacks on several of the problems would require prohibitive amounts of CPU time. Some problems featured very slowly converging series or very large matrices. Other problems were dangerously close or beyond the limits of double precision arithmetic. While the team made heavy use of mathematical software including Maple and Matlab, it was insight and craftiness that transformed the inaccessible into the routine in almost every problem. In the end, all of the team's solutions required at most a few minutes of CPU time.

Back to UD Math Home Page

© 2002, Department of Mathematical Sciences
Last Updated:
Comments? E-Mail Webmaster(www@math.udel.edu)