What would be a good way to immerse myself into programming A.I.? (1 Viewer)

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Fairly new to Computer Science at University. I am interesting in Machine Learning and Artificial Intelligence. Although, I am unsure as to where I should start. Any suggested reading? courses? Programming Languages or any advice?

Thank You
 
Your university should have a few courses on it. Do you have solid programming background or are a beginner?
 
A beginner I know scripting languages fairly well like Javascript, HTML but other than I am a novice.
 
You'll need a very high level of programming experience for ML and AI. You'll need to start working through programming books. The popular languages are C++, JAVA and Python. You'll also need quite a bit of math. These fields are very demanding.
 
I'd like to understand them better, it's endless fascinating. Just to clarify in what ways are they demanding exactly?
 

.Scott

Homework Helper
2,160
681
What would be a good way to immerse myself into programming A.I.?

Hmmm, trying to get yourself to the technological singularity? Ultimately that would require s team of system engineers, EE's, and software people - all, presumably, ready to retire very soon.

Most of what's in AI can be understood by any programmer using a general purposes language. The HTML/Javascript combination can do a lot - but you probably want to get some C++ under your belt.

The classical AI experiments were all done decades ago - and with machines that can now be matched with Javascript. For example, write a Javascript Tic Tac Toe problem that keeps track of it's wins and loses so that its game improves based on experience. Javascript could handle that.

Of course, once you've figured out how to take over the world, you're going to want a more powerful language - perhaps C++ or perhaps you'll have an array of custom engines that you'll need to program from scratch. In that case, you'll need to learn about macro assemblers. The can be configured to program your new machine.

Have fun. If we don't hear from you again, we'll understand - it's the immersion.
 
615
35
Fairly new to Computer Science at University. I am interesting in Machine Learning and Artificial Intelligence. Although, I am unsure as to where I should start. Any suggested reading? courses? Programming Languages or any advice?

Thank You
In all seriousness, I would start here: http://en.wikipedia.org/wiki/Machine_ethics. I'm sure it's not the area of AI that you're immediately interested in, but you will happen on it eventually and it's the area most likely to shape what you want to learn, so it's a natural choice for a starting point.
 

AlephZero

Science Advisor
Homework Helper
6,920
290
In all seriousness, I would start here: http://en.wikipedia.org/wiki/Machine_ethics.
Ethics is a branch of philosophy, and to a first approximation western philosophers are still having inconclusive arguments about the same topics that were done to death by Aristotle and Plato.

If you want to learn about AI and MI, then learn about AI and MI. Ethics is irrelevant until you actually produce a working system that does something significant.
 

.Scott

Homework Helper
2,160
681
In all seriousness, I would start here: http://en.wikipedia.org/wiki/Machine_ethics.
I'm with AlephZero on this one. I read that wiki article - it's pretty vacuous.
But there's a more important point. If you're programming a machine that has the mechanical ability to do serious damage or even kill, then there is a safety issue. Dealing with that safety issue as a ethical issue means that you're presuming that the software will operate as you intend and so restricting is with software-implemented ethical limitations will prevent it from acting unsafely. In fact, depending on such elaborate software as the primary safety constraint would be monumentally unsafe and unethical.
 
Google "Game Theory", there are some great stuff that can help for building an A.I.

It does sound like it only applies to games but it applies to a lot related topics which you might find fascinating.
 
615
35
If you want to learn about AI and MI, then learn about AI and MI. Ethics is irrelevant until you actually produce a working system that does something significant.
For obvious reasons the very compelling argument is that is exactly the wrong order in which to approach AI.

I'm with AlephZero on this one. I read that wiki article - it's pretty vacuous.
Yeah, the wikipedia article could do with some work, but I offered it more as a starting point than a good reference for the subject matter.

Perhaps, this serves that purpose: http://www.springer.com/engineering/computational+intelligence+and+complexity/book/978-3-642-32559-5
 
Last edited:

The Physics Forums Way

We Value Quality
• Topics based on mainstream science
• Proper English grammar and spelling
We Value Civility
• Positive and compassionate attitudes
• Patience while debating
We Value Productivity
• Disciplined to remain on-topic
• Recognition of own weaknesses
• Solo and co-op problem solving
Top