Mathematics for a programmer

Question

Mathematics

Hello. I’ll be brief:
I’m weak with mathematics, and for a programmer this is unforgivable, so I decided to improve - please tell me the best way so as not to get bored and not stretch the training for years.
Thank you.

Answer 1

> for the programmer it is unforgivable
Small offtopic. For 90% of programmers of the 9th grade level, it's enough. I foresee the exclamations of representatives of the ilita, they say, you need to learn mathematics if you do not want to be the coding monkey of blablabla. The topic “does a programmer need mathematics” is debatable, you can use hardcore mathematics for a penny in some kind of research, or you can do boring, highly abstract work of a quota for a salary 2 times higher than the market, in general, mathematics as an end in itself is not very good IMHO.

Answer 2

An abstract answer to an abstract question:
Set theory
Graphs
Trees
Transport problem
Neural networks
Artificial intelligence :)
Or what do you need in your direct specialty? Models? Numerical methods?

Answer 3

For all my life, only a school mathematics course + 1 university course (where about matrices and determinants) was useful to me when working with graphics.

Answer 4

Effectively mastering a large amount of information in a short period of time is, to put it mildly, not an easy task.
If everything is in order with self-discipline, then you take books from the course on higher and discrete mathematics and other related disciplines for technical universities and study step by step.
If not, then perhaps it makes sense to enroll in the correspondence department of a technical university.
These thoughts are given for a spherical programmer in a vacuum who is weak with mathematics. But in my opinion, this will give the most complete and high-quality training.

Answer 5

It's never too late to learn. I would pay attention to the following sections: probability theory and statistics, mat. logic, information theory (complexity of algorithms). Then choose what interests you.

Answer 6

I am self-taught, there was no discrete mathematics on Instagram either. Now I felt that it is simply necessary to improve my knowledge of mathematics, I bought a textbook, I delve into it.
If there is good motivation, then you can quickly figure it out on your own, the main thing is to understand where, what and how is applied in practice.
I can recommend this book www.ozon.ru/context/detail/id/3510820/. Written relatively accessible, but there is no theory of algorithms, theory of automata (lectures on them are not a problem to find). If not suitable, look here habrahabr.ru/qa/1856/

Answer 7

Graph theory and combinatorics (sections of discrete mathematics) are very common. I advise you to pay attention to it.

Answer 8

Yes, okay you. I am an Applied Mathematician by education, for 5 years its various sections were the main subjects, but still at times I feel that I don’t know so much mathematics (for example, when I read books on cryptography).

Answer 9

1) Set priority goals for yourself - what you want to learn first. The rest is better not to learn (and if you teach, then do not go deep, because it is impossible to learn everything)
2) Buy / find literature on your topics. Books are not only theory, but also a huge pile of practice.
3) If it's hard to learn on your own, hire a good tutor.
4) Set yourself a task like: "Read a paragraph of theory, solve 15 problems on it." Moreover, the tasks should be fairly simple at first, and complex at the end. It is clear that problems cannot be left unsolved (that is, if you cannot solve it, look at the solution).

Answer 10

when you mean math for a programmer, do you mean algorithms + calculus + discrete math?
then a standard set of books:
Kormen, Knut (at least a few volumes), Romanovsky.
there is a rather informative site intuit.ru , on which, with good self-discipline, you can squeeze out a lot of knowledge.

Answer 11

Depends on what you are doing. For example, system programming needs mathematical logic, theory of information and algorithms, probability theory, algorithms from Knuth, number theory. If pattern recognition, then probability theory, functional analysis is real and complex. If symbolic calculus, then algebra, representation theory. In general, it depends on the place where your talents are applied. If modeling, then numerical methods.
There is no common denominator here, I guess. Although, from experience, probability theory and information theory are always useful, and the basic algorithms from Wiener's book 'Algorithms and Data Structures'.

Answer 12

Discrete Mathematics for a Programmer
There are graphs, sets, relations, combinatorics and all the necessary minimum of discrete mathematics.
And from the highest - the main thing is to understand the basics, and even if you encounter the task of decomposition and correction - this is only partially higher mathematics - probability theory studies correlation, statistics and random variables. And the series, differentiations (if necessary) can be googled and remembered, kept in mind constantly and refreshing the memory with this information is not worth it - this is very often used in real conditions in programming. But there are a lot of sections of mathematics and the overwhelming aspect of knowledge of mathematics is applied very conditionally - from case to case, since working as a programmer is not working at a research institute, and mathematics will already be needed where nothing can be solved without it) And mostly linear more than enough algebra.