Self-studying plan for modern science

[Angelo422]

Hey guys, I want to build a strong and straight plan for my next years of studying and once finish I am able to do something on my own and come up with crazy ideas and actually test them, build some awesome algorithms, all that cool stuff, but I'm kinda stumble so it would be nice if someone could help me with that.

Hopefully this will be a kind of a guide for those that are with the same issue that me.

Just a heads up:
- Assume that I don't know exactly what I'm talking about because everything I will talk about is my own research, so I might miss something, say something wrong or just something that don't quite heads up (I don't know anyone to talk about this topics so it's hard to organize all this by myself)

I know this will sound kinda stupid, vage and sort a pipe dream from a guy that doesn't even know what linear algebra is about, but my final goal is to study the "hot topics" in science like Machine Learning, A.I, Quantum Physics, adding them up and, of course, I want to explore the space out of this planet (not literally, just want to be apart of the revolution (this is the most vague idea that I have but is the one that gives me the most ambition to follow this - I always though physics was to much for me, and yet I loved Mat. -))

As Picasso said: "Learn the rules like a pro, so you can break them like an artist". So I can't jump right into quantum physics, I need to know the fundamentals in order to understand how things work and establish a connection between both (again apologizes for my ignorant way to talk about this, I know this goes way futher and my way to talk about this probably gives you laughs). But as I said, I want to know ML and AI, and honestly starting with physics etc doesn't seem the right way to go, so I might learn ML, do some projects, maybe go to AI (I don't know much about this topic in terms of study and requirements) and do a project aswell. (In the beggining of all this I thought a good way to start was with computer science but a few weeks ago I started getting cold feet, because such course will take a lot of time, as well afford (since I will be learning by myself with the help of the internet :'D) and is too abstract, doesn't apply right the way (like ML) and focus a lot in computing unlike ML, Quantum physics and a whole lot of things that rely more on Math, etc and such a little percentage of computing.)

This as far as my plans go for now, I'm completely lost and I could use some help make my plans to get to my goal..

Thanks to all of you that took a little of you time to read this and I'm even more thankful to those that will reply.

 

[berkeman]

Welcome to the PF, Angelo.

Absolutely learning mainstream science is a necessary first step toward trying to think outside of the box. A solid foundation in math and science is super important.

Where are you right now? Have you graduated high school? If so, what were your highest level of science and math courses? Did you enjoy them?

Have you looked at the online learning resources that are at the beginner undergraduate level? Courses like the Khan Academy and the MIT Online Lectures?

 

[archaic]

I know this will sound kinda stupid, vage and sort a pipe dream from a guy

We indirectly influence ourselves, don't say or write such things. If it sounds stupid for a chap it wouldn't for another. Remember that, at a time, it sounded hilarious when someone said the Earth wasn't flat.

 

[DEvens]

It's only my suggestion, but I would suggest you pick a particular area and learn that as well as you can.

If you are interested and motivated in a topic then it will be a lot easier to keep yourself working at it. If you have something that's a chore and a bother and tedious, it will be very tough to keep going back to it on your own motivation.

No matter what subject, it is very unlikely you will find any really interesting results "at the door" so to speak. You will need to grind through the basics. For most subjects, for ordinarily bright people, this takes 4 years of an undergrad degree. It's possible to get through a main-line of a subject in less. But do remember the Dunning-Kruger effect.

https://en.wikipedia.org/wiki/Dunning–Kruger_effect
So machine learning and artificial intelligence work together. It's possible you could do something useful in this area. If you want quantum mechanics then maybe you want quantum computing. This isn't specifically about ML or AI. Quantum computing is much more about solving specific kinds of mathematical problems, not about solving a much larger and broader problem such as ML or AI.

So to come back: Pick a topic that really energizes you. Then go get the intro texts. There are tons of questions posted on here about "what's the best text in..." so hunt around here. And Google and Amazon for the good books.

Much of physics rests on math of various kinds. Probably you want to start with calculus, and look around for other things that look useful. Complex analysis, algebra, geometry, etc. Many parts of physics use other forms of math, some quite familiar like statistics, other forms less familiar. One day I hope to really learn what is a quaternion, for example. Group theory will probably be useful. If you go into gravity at all you probably want some differential geometry, maybe some topology.

But each of those subjects is potentially an entire life's work on its own. So it is important to try to pick a topic you want to get good at and concentrate.

 

[Angelo422]

Welcome to the PF, Angelo.

Absolutely learning mainstream science is a necessary first step toward trying to think outside of the box. A solid foundation in math and science is super important.

Where are you right now? Have you graduated high school? If so, what were your highest level of science and math courses? Did you enjoy them?

Have you looked at the online learning resources that are at the beginner undergraduate level? Courses like the Khan Academy and the MIT Online Lectures?

Thanks

Yes, I've just graduated high school but it was in graphic design, we just had math and it was kid's level. Never did any "advanced" course but I have no doubt that I'll enjoy.

In matter fact I have. My plan for ML includes Khan Academy, MIT, edX and some youtubers, that's probably the resources that I'll use to learn everything.

 

[Angelo422]

We indirectly influence ourselves, don't say or write such things. If it sounds stupid for a chap it wouldn't for another. Remember that, at a time, it sounded hilarious when someone said the Earth wasn't flat.

I'll keep that in mind, thanks!

 

[Angelo422]

It's only my suggestion, but I would suggest you pick a particular area and learn that as well as you can.

If you are interested and motivated in a topic then it will be a lot easier to keep yourself working at it. If you have something that's a chore and a bother and tedious, it will be very tough to keep going back to it on your own motivation.

Yes, I agree, that's why I'll choose to start ML.

No matter what subject, it is very unlikely you will find any really interesting results "at the door" so to speak. You will need to grind through the basics. For most subjects, for ordinarily bright people, this takes 4 years of an undergrad degree. It's possible to get through a main-line of a subject in less. But do remember the Dunning-Kruger effect.

Of that I've no doubt, I content myself with the little progress that I'll do. And since I'm not in any course or school I might finish early. But I won't be thinking anything specific in terms of what I will be, how good I'll be or how fast can I be. Delusion doesn't do any good to anyone.

So machine learning and artificial intelligence work together. It's possible you could do something useful in this area. If you want quantum mechanics then maybe you want quantum computing. This isn't specifically about ML or AI. Quantum computing is much more about solving specific kinds of mathematical problems, not about solving a much larger and broader problem such as ML or AI.

I didn't meant that. I know that they are diferent subjects but I'm talking about a big plan, imagine that I know AI good enough and then I learn quantum mechanics and after that (or between) I put quantum computing to do AI. Like "the neural qubit" - it's a research paper.

So to come back: Pick a topic that really energizes you. Then go get the intro texts. There are tons of questions posted on here about "what's the best text in..." so hunt around here. And Google and Amazon for the good books.

Much of physics rests on math of various kinds. Probably you want to start with calculus, and look around for other things that look useful. Complex analysis, algebra, geometry, etc. Many parts of physics use other forms of math, some quite familiar like statistics, other forms less familiar. One day I hope to really learn what is a quaternion, for example. Group theory will probably be useful. If you go into gravity at all you probably want some differential geometry, maybe some topology.

But each of those subjects is potentially an entire life's work on its own. So it is important to try to pick a topic you want to get good at and concentrate.

Will definitely do that! Yes there are really awesome books out there!

Thanks a lot, will lot into all of that and focus on what I want in order to finish my studies before I die . But yes definitely learn something that I feel interested about.