1
$\begingroup$

I know this is an opinion-based question and will be closed but this is the only place I know that can answer it reasonably and it is a very important matter to me.

I am pursuing a machine intelligence track at my university and as I am moving along from cs fundamentals to track courses I am stuck on what math to take first: Linear Algebra or Statistics. I know both are essential pillars and are interdependent, but I am leaning towards taking stats before linear, which is the opposite of what most ppl do here (hence the hesitation).

Thank you and sorry for making it opinionated.

$\endgroup$
  • $\begingroup$ what are the ML classes that you will be taking in parallel? $\endgroup$ – oW_ Oct 31 '18 at 22:53
2
$\begingroup$

Not sure how intense your professor is going to make either course, but assuming it's the hardest possible introductory course, it would be better to take linear algebra before statistics. A lot of statistical operations require fundamental linear algebra concepts. You need to be able to understand abstract notions like vector spaces and fields to know when your data is in the correct range, you need to be able to understand what operations are linear and what are non-linear so you don't apply a wrong operation on something, and you need to most importantly be able to understand linearity and non-linearity because that is the basis of a lot of statistical methods. Statistics involves transforming data in a way that makes sense but you learn the fundamentals of transformation in linear algebra. But if your stats course is just about p-values and superficial hypothesis testings, and your linear algebra course is just direct applications like row-reducing matrices, then it doesn't really make a difference which you take first.

$\endgroup$
1
$\begingroup$

Can you comment on why this is a very important matter to you? I'm a Ph.D. Candidate in ML and honestly (my opinion, but it is an opinion-based question) I don't think it should be. If you're talking about a first-semester stats course and a first-semester linear algebra course, you'll be fine either way.

$\endgroup$
  • $\begingroup$ These would be intro courses in the sense that I haven't taken then before but they aren't first semester. Linear has calc 3 as a prereq and stats is 300 level (so sophomore usually). I find it important because in parallel I'm beginning to take ml courses and am concerned that by messing up the order of math classes I'll have a poorer understanding of the ml ml coursework $\endgroup$ – mdrjjn Oct 31 '18 at 17:46
  • $\begingroup$ OK, that's a fair concern actually. If there's a 'normal' way it's usually done in the department there's probably a reason for that, so it's probably not a bad default. I suppose it depends on the ML courses you're taking in parallel: deep learning and you're probably better off with Linear Algebra; some more 'traditional' algorithms you'd probably be better off with statistics. But again, it's close - don't let yourself get overly stressed about this decision. Talk to an advisor at your institution if you're really worried about it, they'll know the specifics of the program. $\endgroup$ – Matthew Oct 31 '18 at 17:55
0
$\begingroup$

If you will only be a user of ML apps/high level api-"select from menu" apis (MS Azure etc) STAT will be OK

If you will deep dive and develop algorithms, definitely you would need linear algebra where you can learn matrix, matrix multiplications, linear equations, vector spaces

In many ways, Linear algebra is the fundamental

$\endgroup$
0
$\begingroup$

For beginners stat is much more important, especially for practice. Stat helps understand metrics much better, especially for regression problems and that is what matters the most in practice.

As for LA, understanding basic concepts like vectors, matrices, their addition, multiplication is enough.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.