Skip to content

Latest commit

 

History

History
40 lines (24 loc) · 3.83 KB

BAS.md

File metadata and controls

40 lines (24 loc) · 3.83 KB

← Back to overview

How To Learn & Basics

Are you asking yourself "How do I start in Artificial Intelligence, Machine Learning or Deep Learning"?

In short, just iterate three steps mentioned here until you'll become professional or you'll get your Nobel Price in AI. For more info check out an article how to start learning AI&DL.

Step 1: Math and Programming

For even being able to start with, there's a necessity of having some background in mathematics (such as linear algebra, probability, statistics, and calculus) and programming. Both are essential for AI, ML or DL.

Step 2: Online Classes

Online classes are the simplest source of knowledge on the internet. Check out more in subsections in this Knowledge Base, such as in Machine Learning and Deep Learning.

Step 3: Solve a Problem

Find a problem concerning this topic to play with and solve it. If you have no idea what problem to find, try Kaggle.

Frequently Asked Questions

The difference between

Artificial Intelligence and Machine Learning

In short Machine Learning (ML) is a sub-field of Artificial Intelligence (AI). AI is the intelligence displayed by machines in contrast to natural one in humans or animals. In a general usage sense, AI is "anything that computers can do." ML is a specific set of techniques that gives computers the ability to learn without being explicitly programmed. It usually involves big batches of data and finding ways how to train or code an "intelligent system" that will process such data.

Machine Learning and Deep Learning

Many people would lead you to believe it is a matter of opinion. But by definition Deep Learning should be part of a broader family of machine learning methods based on learning data representations, as opposed to task-specific algorithms.

I'm Bad at Math or Programming. Can I still learn it?

Mostly you can tag along, but at a certain point, if you don't know the underlying Math, you won't be able to understand what you are doing. Same for programming, if you never implement one, or trace one yourself, you will never truly understand why an algorithm behaves a certain way.

So what if you feel you are bad at Math? Don't beat yourself too much. Take Barbara Oakley's class on Learning How to Learn. You will learn more about difficult subjects such as Mathematics, Physics, and Programming.

Recommended Programming Language

In general, if you are working on a computer, you will need to switch between languages from time to time. So arguing about which one is better is not a wise thing. If we want to give the best advice, perhaps the answer is you should just learn as many of them as possible.

Of course, your time is limited. So AIDL usually recommends Python as your first choice. Python is cited as a useful language for AI/DL because it has the best support of libraries. Most ML libraries from python links with C/C++. So you get the best of both flexibility and speed.

Other also cites Java, Lua, Lisp, Golang or R. It depends on your purpose. Practical concerns such as code integration, your familiarity with a language usually dictate your choice. R deserves special mention because it is popular in some brother fields such as data science.

Finally, there is Matlab/Octave. Many Coursera ML's programming is Matlab-based, for example Andrew Ng's class and Geoffrey Hinton's class. You can finish them in Python, but you will have to do a lot of extra work.