Chris Williams' Information for Prospective Students
I expect to take on about one to two PhD students per year.
See my research interests page for
information. I am also potentially interested in
supervising good students who propose their own topics,
as long as these generally fit under the area of probabilistic machine
To work in my group the ideal preparation would be
- Strong preparation in continuous mathematics: linear algebra,
calculus, optimization in continuous spaces
- Strong preparation in probability and statistics: basic
probability theory, probabilistic graphical models
(e.g. Bayesian networks)
- Strong programming background and experience in C/C++ and/or
python and/or MATLAB etc
As a general policy, I encourage potential PhD students to first study
for an MSc degree, as this allows them to gain the necessary
background before embarking on research. If you do not have the
background mentioned above a good way to obtain it is through the
in the School of Informatics, and in particular the
Learning specialism. Normally admission to this specialism will
require a good degree in a numerate discipline.
Informal enquiries about PhDs may be made to me via email (see
my contact information page).
When doing this please provide
Please use .pdf or .txt formats, not .doc.
Further information is available from the
School of Informatics
Postgraduate Prospectus, including information on
how to make a formal application.
- A CV (resume)
- A list of courses taken and grades obtained
- A 1-2 page statement of research interests
You can find information about the
programme in the School of Informatics, and in particular
Learning specialism by following these links.
Normally admission to the Machine Learning specialism will
require a good degree in a numerate discipline. The application
procedure is described
There is also a
in Data Science which has similar entrance requirements.
The School of Informatics (SoI) does not run an internship
program. Please do not email me with requests for
internships -- these will simply be ignored.
Last modified: 9 Nov 2017