If you wish to do Honors in the general area of Nonlinear Optimization/Scientific Computing under my direction, it is essential to arrange an appointment with me during the spring before your senior year. Your senior year will fly by amazingly quickly, so to do a decent Honors project requires some advance planning both on your part and on mine.
Even better would be to meet with me no later than the fall of your junior year with the idea of having you spend the summer before your senior year getting a head start on your Honors project. With sufficient lead time, we can apply for student research grants through the Charles Center to support your research over the summer as well as arrange for campus housing for the summer. In addition, we can discuss possible topics and consider the courses, etc. that it might be advisable for you to take during your junior year in preparation for your Honors project.
For instance, in most cases it would be advisable for you to take CSCI 315, Systems Programming no later than the Spring semester of your junior year. Also, it would be a good idea to take at least the first of the two numerical analysis courses in the Department of Mathematics (MATH 413 and MATH 414, both of which can be counted toward a degree in Computer Science) during the Fall/Spring semester(s) of your junior year. Finally, I would urge you to consider taking CSCI 426, Simulation, by the Fall semester of your junior year, though if you have not done so, it would be advisable to take CSCI 426 the Fall semester of your senior year.
Students working with me on Honors do a great deal of experimental work in floating point arithmetic. To effect extensive experiments in a relatively painless fashion requires adept systems programming. To understand the perils and pitfalls inherent in floating point arithmetic requires understanding at least some of the fundamental issues introduced during the first semester of the numerical analysis course sequence. Finally, since the experiments usually make use of pseudo-random number generation to generate data or initial parameter settings, it is valuable to have the simulation course since it makes use of a pseudo-random number generator within the context of careful experimentation.
Last updated Saturday, 3-Aug-2013.
Copyright © Virginia Torczon. All Rights Reserved.
The Department of Computer Science
Arts & Sciences Graduate Program
The College of William & Mary
William & Mary Events