Applied mathematics as a second degree for a programmer?

Question

IT education

Good afternoon. I am 26 and I have a degree in programming, I have been working in my specialty for 6 years, I successfully got a job working from home, I have free time, and now I decided to think about a second higher education. I didn’t have to choose the area for a long time, from the very beginning I wanted to enter the PMA, but at the last moment I went to programming. It makes absolutely no difference to me if I have a crust (I myself was expelled from the 4th year, so as not to be distracted from work, because I understood that it was of no use at my university, though then, under pressure from my parents, I recovered). I am interested in the specialty itself (namely, mathematical modeling), and related areas of programming (it’s clear that everything is related there, but I’m interested in a deeper understanding of the theory of algorithms, machine learning and distributed systems from a mathematical point of view). But basically, I'm just interested in methods of mathematical modeling of both economic / physical processes, and (mostly) and the application of applied mathematics to programming. As an option, further development of big data is possible (although judging by the book by D. Foreman, this is the 4th year of the university at the programming department). And in general, I want to better understand my specialty and be able to, say, fully understand Knuth (his mate calculations are put into a stupor).
What do you advise? Is it worth doing? Will they take it if you are only a specialist, and even expelled? Isn't it too late? Are there alternatives? Is it possible to attend lectures freely and is it better?

Answer 1

There will not be a correct answer here that will be absolutely correct. Here, how can you answer "yes, do it, because this is the only way you can get a complete and structured knowledge base", and "no, why once again waste time on incomprehensible training of everything and everyone, it's better to get yourself normal books, and sit and teach yourself\ solve \ prove, etc.". So I'll just share my idea in this matter.
My personal IMHO, which can be both right for someone and fundamentally wrong: it’s worth doing, but (!) if you are sure about the university you plan to enter, or rather, in the department you want to go to!
I'll start from the distant: mathematics is "the queen of all queens." And yes, even though I'm only in my second year of university, I've only recently REALLY encountered the fact that so many seasoned programmers repeat so often on various smart forms: "you need a matan, you love it!". And seeing all those disciplines that were named above, I can say for sure that there is no way without a university.
Let me explain by my own example - I entered a university, which is considered the best technical university in the country (KPI, hello). But he did not enter a very popular faculty, about which only a deaf-mute had not heard, but a faculty. little-known, but by profession - computer science.
Then, like everyone else, I had a "regular" course in higher mathematics (3 semesters), a semester course in the theory of algorithms, linear algebra \ computational geom and discreet, as well as a semester in probability theory. I, like all conscientious students, attended lectures, solved problems in practice, studied theorems with proofs (fortunately, it was easy, because from the physics and mathematics) and didn’t really understand where all this could come in handy for me (I’ll clarify, the course on theoretical algorithms was rather weak, so I cited it as an example with other subjects purely for statistics). And I lived so-so calmly, and would continue to live without thinking about the role of the matan in my life, when by chance one of my teachers pointed me to his colleague, who began to teach such a subject as "Structural and static pattern recognition".
Well, after the first lecture and then another practice, to say that I "saw the light" and understood why I need all that mathematical base - to say nothing! Moreover, the further I studied this subject, or rather, the subject area, the more I understood that "what a pity that I entered the faculty, where mathematics is so weak."
And indeed, communicating with other guys from physics and technology. faculty with which I go to this subject, I realized that if I still want to stay in the "topic" and continue to engage in recognition, then in addition to all my knowledge, I will also have to "catch up on lost time", i.e. to start studying those sections and topics in math.analysis\set theory\disk.mathematics\analyst.geometry etc, which I simply didn’t have!
But before summing up, I want to make a reservation right away: everything is possible, and if you have enough motivation, strength and everything to master at least ~ about the first two years of a good university course specializing in mathematics (or so) - then you definitely don’t need a university, a list of references will suffice. After studying all of the above material, in the future, a calm entrance will be provided in areas such as machine learning or pattern recognition! But to be honest, I have never met such people. All my acquaintances and friends first of all studied these subjects at the university, and after that they started doing something more "interesting" :)
Therefore, based on my personal experience, I would advise you not to be lazy and still do it! It's never too late to learn! The main desire!

Answer 2

That is, you want to say that you are ready to jump headlong into esoteric theoretical computer science (such as cstheory.stackexchange.com ), which is absolutely not applicable in 100% applied tasks and freelancing. Which is used by 3-4 companies throughout the CIS (Epam, Google, Yandex, Mailru)? You will sit for six months without work, because. accidentally made a mistake in solving a math olympiad problem for grade 8, but in php and java they take it without all this?

Answer 3

If you wish, you can master everything on your own (textbooks, lectures, courses and resources on the network in bulk). As for interaction with a teacher, who supposedly can be asked something, then there are no problems with this on the Internet, there is math.stackexchange.com , for example. However, this path is simple only in words, in practice it requires a great organization and self-control. However, it is possible .
And now about full-time study at a university (I don’t know anything about part-time education)
If there is a desire and an opportunity, then why not? Some branches of mathematics like discrete mathematics, linear algebra, probability theory and statistics are quite useful for a programmer. Especially in the field of data analysis, for example.
On the other hand, not all mathematics will be useful. I will not give specific sections, because it depends on what you will end up doing. However, it is unlikely that all disciplines studied will be equally useful. Hence the feeling that mathematics has been studied much more than necessary, which is partly true - in order to consolidate the base well, you need not only solve problems, but also understand how it interacts with other areas.
If you are interested in applying mathematics to programming, then I advise you to focus on programs that explicitly mention computer science / computer science. Of course, any applied mathematics is currently done using computers, but somewhere it is just a tool for calculating, say, a complex differential equation, and somewhere mathematics, on the contrary, is a tool for solving a programming problem (recommendations in an online store, for example ).
Well, in general, a lot depends on the university. I would not advise going to the first one that came across, but there are few decent ones, most of them in Moscow / St. Petersburg.

Will they take it if you are only a specialist, and even expelled?

And what problems can there be? As far as I know, in Russia only one VO is paid for by the state, i.e. have to pay for education. Well, since you are already paying, I see no reason not to take you :-)
And there is also such a loophole: the Russian system of higher education is now slowly switching to the Bologna system of education, expressed in 4 years of bachelor's degree + 2 years of master's degree. As far as I know, specialists from previous years of graduation have the right to enter the master's program on a budgetary basis. It seems that such a scheme is even better: 4 years is a fairly significant investment of time, and 2 is already more real.

Is it possible to attend lectures freely and is it better?

Just listening to lectures is useless a little more than completely. We need to solve problems, without it in any way.

Answer 4

Thanks for answers. Yes, to be honest, I don’t really trust our education system myself, and personally it would be much more convenient for me to study on my own - and I generally do that - I watch courses, I use mathematica as software. The problem is that I very often lack live communication with a teacher who would explain things that are not clear to me. Of course, math.stackexchange.com can help with this, but by their very nature, stackexchanges are aimed at specific questions, and if you are a noob and want to understand some more basic things, and even formulate the question not quite correctly, you will most likely be downvoted.
I'm thinking it would be cool, let's say, to find some kind of sensible teacher, and pay him for consultations, say, once a week for an hour or two. Like a tutor in applied mathematics and computer science. I wonder if this is possible?)

Answer 5

Education in our country is very far behind. (Except for a couple of universities).
So if you want to really learn something, then you have to go to Europe or the states to study.
In Germany, for example, you will find a job and be able to get an excellent education. Yes, you travel.
Keep it up while you're young It's
never too late to learn. People in their 50s and 60s learn and relearn)