Ask HN: What's the best way to relearn math?

62 points by etr71115 ↗ HN
Recently, I've noticed my quantitative abilities slipping. I want to refresh myself on probability, statistics, linear algebra, and maybe calculus?

Piggybacking on the thread below, I'd be curious to hear any new sites/apps/books/practices the HN community recommends. This is more of a mental exercise than an application towards a particular problem. Thanks in advance!

https://news.ycombinator.com/item?id=3569761

20 comments

[ 3.0 ms ] story [ 62.6 ms ] thread
Install Anaconda. Start playing with open science data sets. Gain superpowers. Solve big problems. Good luck!

https://www.continuum.io/downloads

This might be a slightly better page: https://www.continuum.io/anaconda-overview

But, they really don't give enough info about what, where, why. What data? Who is this intended for? Etc.

I think the poster assumed that data science/computational work could provide the mathematical background the OP is looking for.

Because Anaconda is a package of tools that include numpy,scipy (python numerical and scientific computing) and a notebook (a "live editing" utility).

While this might help, I wonder if this is too roundabout a way to learn math.

Same boat. I was just going through boxes in my garage and came upon my old college math textbooks and some old exams/homework assignments. It was like gibberish.

My plan was just to take a textbook and see if I can just relearn. Probably are better options out there, but I'm clueless.

Probably will start with either linear algebra or maybe set theory?

Happy to hear other ideas of how people would do this. For me, there's no practical application. I just used to love this stuff and hate how much I've forgotten.

> Probably will start with either linear algebra or maybe set theory?

This can work. I spent a year or so before grad school at work, I brushed up by spending a summer drilling through a Linear Algebra book.

A new thing I have been trying out recently is find an application, find something I am confused by, dig deeper through the layers of mathematics till I uncover the basics of the confusion, then re-learn and work up.

So BFS vs DFS. The latter seems more like up my alley. YMMV

I am the main developer of the MathPiper computer algebra system (http://mathpiper.org). I found that a great way to improve one's understanding of math is to learn how a computer algebra system works. I would be happy to teach the fundamentals of a CAS to anyone who is interested.
I have started taking classes at my local community college to refresh my math skills again.
For Algebra, if you need certain subjects in depth or need to take a test to assess your skills, try Khan Academy [1]. Also, Khan covers trigonometry and geometry. For Calculus, MIT did a series called Calculus Revisited which covers calc 1,2,3.[2] Calculus Revisited was designed for professionals who needed a refresher or first crash course. It's from 1968-72 but I thought they were great.

[1] https://khanacademy.org [2] https://ocw.mit.edu/resources/#Mathematics

You can go through Khan Academy and I find MathematicalMonk's lectures on youtube very easy to follow.
Even mental exercises could use a purpose. What are you interested in? Try to keep an end goal in mind. If you want to learn more about machine learning, it might be an idea to start with a mathematical preliminaries chapter in a typical machine learning textbook and look up difficult topics in a probability/statistics textbook. If you want to learn more about databases, perhaps you could pick up a book on relational algebra, and explore it in the context of abstract algebra. There are many ways to get the job done.
When I was preparing to apply to grad school ~4 years out of undergrad I spent a lot of time working through the various Math modules on Khan Academy and found it very helpful for getting me back up to speed on the basics. They have some higher-level math content on there too now, I believe.
Here are some excellent YouTube playlists for the subjects you mentioned. I have used and can vouch for the high quality of all of them:

Linear Algebra: https://www.youtube.com/playlist?list=PLlXfTHzgMRUKXD88IdzS1...

Probability: https://www.youtube.com/watch?v=KbB0FjPg0mw&list=PL2SOU6wwxB...

Calculus: https://www.youtube.com/user/professorleonard57/playlists

Be sure to work through problems and not just watch lectures, because otherwise you'll forget the content. For example, the linear algebra course has a companion site with problems, and the professor of the probability course wrote an excellent book with problems.

I've been doing Khanacademy. Its been fun.
I'd love to have enough time to read everything about homotopy type theory / category theory / number theory / cryptography. Heck I'd even want to read the complete Art of Computer Programming from beginning to end. It's so hard to balance this with being productive. Life is too short.
I asked this question on reddit a few days ago!

My basis was I wanted something like Charles Petzold's "Code" from 1995 which is the BEST way to understand computers.

The recommended book by far was..."Mathematics: From the Birth of Numbers".

See here: https://www.reddit.com/r/math/comments/673yxa/best_book_for_...

I'm reading Code right now. It really is a fantastic read.
I have found http://www.betterexplained.com very fascinating (no affiliation).

The author, a person also interested in programming, comes up with intuitive ways to explain mathematical concepts.

Khan academy and youtube in combination are probably better then any of the schooling I originally recieved