While numerical methods predate the invention of both mechanical and electronic computers, their ubiquity in modern physics only established itself with cheaply available personal computers. Today, almost all physicists, including theorists and experimentalists can benefit from computational skills.
Like much practical computer science, numerical methods are particularly prone towards self-teaching and independent experimentation. This is largely because a wealth of readily accessible and comprehensible information exists online regarding the techniques and algorithms one may wish to apply. Learning how to research and implement these algorithms independently is superior to trying to memorize specific techniques. With that in mind, we have tried to overview some general key topics with only a few details and references, leaving the actual algorithmic research and implementation up to you.
The fall term will focus on Classical simulations. The winter term will focus on a variety of other topics, including Monte-Carlo simulation, symbolic algebra, and random matrices.
Instructor: David Simmons-Duffin, Lauritsen 442, email: dsd.
TA (fall): Brenden Roberts, email: broberts.
Offered: Fall and winter terms, 2018-2019.
Class meetings: This is a project-based course. There are no official lectures. Lab hours will be set by a poll at the beginning of the term. If you are enrolled in or auditing the course, please join the moodle to ensure you get announcements.
Grading and homework: This course is pass/fail. There will be four projects over the course of the term made available on the course moodle.
Prerequisites: Students should be comfortable with the command line and familiar with at least one of Python/C/C++.