knowLittle said:
Hello, I'm a CS major.
What should I study as a minor or maybe major to study Artificial Intelligence?
Hey knowLittle.
My recommendation is to at the very minimum study probability and statistics.
The thing you need to be aware of is that many AI methods will work in a context of uncertainty: in other words you will be taking data that represents erratic and random bevahiour and you are trying to find patterns in this environment.
If you want an example think about how we humans are able to recognize say the letter A when we read hundreds of people's handwriting that is all different. If it is not typed by a computer, then chances are we will have a hundred different A's but never the less we are able to look at these hundred different A's and recognize it as the same object.
This is the kind of thing that needs to be considered in AI because you will need to take this kind of data and find a way to classify it like we do when we read handwritten letters. The same kind of thinking applies to trying to recognize patterns in general for making intelligent decisions and is not just limited to things like recognizing letters on a page.
Also I recommend taking a class where integral transforms are used and know how to study general integral transforms in a broader way so that you can get a book or a paper or something else and understand what that particular transform does in the context of a general projection.
You can think of integral transforms as decomposition techniques: they take something and split it into parts in the same way that say a vector in 3D space is decomposed into x, y, and z parts.
I would for this take classes in linear algebra, and some kind of signal analysis course from either the engineering department like an EE one, or from an applied mathematics department. Anything with transforms like Fourier transform, laplace transform, wavelets, or signal analysis (Do this if you get the chance over all of the others) is what you need.
Also in your spare time read about anything to do with structures, in particular languages and things like grammars including BNF structures. Remember that language as a general mechanism is a way to take something and constrain it enough so that the representation can be used for some purpose.
The language used and the structure of the language used produces context: The context is used to make decisions in terms of your Artificial Intelligence. Langauge is something that is so important and when you understand language more deeply you'll understand why its a powerful tool in AI especially with trying to understand completely unstructured data and doing things like analyzing what things like DNA actually mean in terms of a language.
In other words, think about the situation where you are given a book in some alien language and you've got to figure out what it means just from the symbols alone and no other information and you've got an idea of how language is crucial to intelligence in the context of AI.
For language I recommend you pick up a book on compilers for computer science.