I'm seriously considering starting a maths degree in October and was wondering if anyone had any advice. Ultimately I'd like to get into game development and electrical engineering so I thought doing a maths degree would kill two birds with one stone.
What are the career prospects for maths degree graduates? I'm very much computer orientated.
Well if you want to get into some area of electrical engineering as an engineer then I'd have thought that an EEE degree would perhaps be a better choice. I guess perhaps you could go for an maths degree, focus on applied maths modules and/or chuck in a physics module or two (like perhaps electromagnetism) and then take up a relevant masters degree or indeed a dual honours maths and physics degree.
I guess with regards to games development you'd perhaps be better equipped than the typical UK CS grad re: the mathematical side of things though they same is perhaps true of physics and engineering grads vs the typical CS grad (course dependent).
It certainly isn't going to harm your career prospects, if you're doing it with career goals in mind then perhaps bias things more to the applied modules and or combine with stats - that could make you very useful in a lot of careers. Having said that getting the best grade is perhaps the priority so perhaps pick the modules you'll likely enjoy the most unless there is no preference between them then go for the more practical ones.
I want to the do the Maths BSc from the Open University.
This is the syllabus:
http://www.open.ac.uk/courses/qualifications/q31
Yeah. When it comes to EE I want to concentrate on FPGAs and digital signal processing which are both very maths heavy. Plus the advantage of doing a maths degree is it has applications in an awful lot of fields.
You could certainly do something like that and, aside from the compulsory 60 credit module in 2nd year and some of the 1st year stuff (which looks rather basic), just make sure that all your 3rd year modules are relevant. I don't think you need to worry that you'd be spending time studying too much stuff that you have no interest in as you're slightly constrained in your choices there if only picking 4 modules at 3rd year, if anything the dilemma is going to be what to not take.
You mentioned theoretical computer science as an interest, they don't seem to offer a mathematical logic course which might have been interesting there, then again that would be taking up credits that might otherwise be used for a more practical/career orientated course. You mentioned AI (if you really want to get into this then it can involve you needing a post grad qualification, though a maths degree is a great basis for that), the multivariate calculus and linear algebra you'd cover in the 2nd year applied maths module would be needed there, also at 3rd year the optimisation module, the probability module and the mathematical statistics module would all cover things needed for AI/ML too (though that is 3/4 of your 3rd year modules already!). The other useful one to have for AI/ML (in particular for bayesian stats/graphical models) would perhaps be the one covering graph theory and networks (in fact that ties in nicely with algorithms to some extent too). From an EE pov you'd perhaps want to have the 3rd year dynamics module 327 as it seems to go further into Fourier methods etc... (The fluid mechanics module might also be of interest re: engineering in general).
Do you already have a degree though? There are perhaps some other options, firstly you could also take a cut down set of maths modules via the University of London international program (distance learning) or if you live in London via Birkbeck College also part of the University of London albeit offering evening classes rather than distance learning. You could then just focus on a few undergraduate modules that get you some of the stuff you want to cover and build up your mathematical maturity in general. That sort of thing combined with a quantitative undergrad you might already hold from years ago in say CS could get you into a post grad course in say data science, ML, or indeed perhaps some signals processing related MSc - worth checking with admissions tutors if you planned to do something like that later as you could save yourself some time by spending 2 years getting a grad certificate in maths focused just on the areas you need for admission vs 4-6 years getting a degree.
Lastly if it is just for interest/knowledge you could perhaps save yourself some time/money by just self studying. Pretty much all the MOOC providers have data science, ML courses. Coursera and EDX certainly have a signals processing course too. Plenty of MOOCs are dumbed down a bit though relative to full fat university courses. There are sites that would be useful to you re: the sort of maths you'd need for both EE/signal processing and ML.
Firstly if you've not done any maths since your non-maths degree at uni or a-level then you'll perhaps need to cover this:
https://ocw.mit.edu/courses/mathematics/18-01sc-single-variable-calculus-fall-2010/
https://ocw.mit.edu/courses/mathematics/18-02sc-multivariable-calculus-fall-2010/
Then you'll want linear algebra - Gilbert Strang's course is well regarded:
https://ocw.mit.edu/courses/mathematics/18-06-linear-algebra-spring-2010/
If you want a book then get this one:
https://www.amazon.co.uk/Mathematical-Methods-Physics-Engineering-Comprehensive/dp/0521679710
Likewise the following Stanford EE courses would be useful not just for EE but also for ML, the second covers lots of the linear algebra covered above too:
https://see.stanford.edu/Course/EE261
https://see.stanford.edu/Course/EE263
https://see.stanford.edu/Course/EE364A
https://see.stanford.edu/Course/EE364B
For a couple of full fat introductory ML courses you could look at:
https://work.caltech.edu/telecourse.html
https://see.stanford.edu/Course/CS229
for a gentler introduction see:
https://online.stanford.edu/courses/sohs-ystatslearning-statistical-learning-self-paced
You can also make use of some university modules that are available openly albeit not on dedicated open learning sites for example, for a more advanced ML course see:
http://www.cs.cmu.edu/~epxing/Class/10708-14/lecture.html
http://cs231n.stanford.edu
etc..
and indeed for something more practical - check out:
https://www.fast.ai
