Career in Artificial Intelligence

[animboy]

I know for a fact that all the relevant knowledge required to achieve this career, I will have to somehow get on my own. But I want to know which is the best "official" career path for me? Would it be electrical/computer engineering, neuroscience, or something else since my university does not offer specific courses for AI since it is quite a new field.

 

[Travis_King]

Computer science and electrical engineering are the two major ones. Programming is huge, but you've got to build/assemble the circuits as well.
Strangely, philosophy is also a path. Though hardly as employable.

 

[Floid]

It isn't really that new of a field, neural network research began in the 1950s. The problem is that a lot of expectations for AI have always been overblown, which led to a lot of decent research being viewed as failures and in turn ever reducing funding for that research.

But for AI you would be looking for either electric computer engineering or computer science.

 

[AsianSensationK]

I don't know a ton about the field, but I was surprised at how much this area draws on different fields. Statistics is used a lot, specifically in the area of machine learning. It also seems like linguistics comes into play, as well as a little philosophy. But yeah, computer science, computer engineering, and electrical engineering are the 3 most relevant degrees. AI first and foremost seems to be about getting machines to "behave intelligently" and these 3 areas give you what you need to understand the basics of machines.

Also, courses are at certain universities depending on what professors can teach and are comfortable with teaching. My university for example has courses in areas like parallel processing, but nothing on graphics. Because there's a professor at my school who does research in parallel processing and not a professor who does research in computer graphics.

 

[hadsed]

Most definitely computer science, or applied mathematics with some computational aspects and the relevant CS courses (algorithms is a must). One of the most practical fields in CS that is a pretty hot thing right now is machine learning, applied to things like data mining and search. The neural networks ideas as well as computer vision require lots of statistics. So will machine learning. Typically these are areas where you need a graduate degree to be able to apply them effectively, or to get a job doing research in these areas (most likely for a company that deals with large amounts of data).

Here are some of the applications of machine learning, on wikipedia:

http://en.wikipedia.org/wiki/Machine_learning#Applications

 

[chiro]

Hey animboy and welcome to the forums.

In terms of useful knowledge I would recommend calculus, discrete math, a good solid statistics pathway, a good solid computer science pathway, and on top of these, any AI specific courses.

In terms of learning algorithms in AI, many are based on probability measures and information, and the goal is to produce a model of some phenomenon that converges to the actual model represented by the data. There are formal definitions for what convergence is mathematically, but that is the basic idea.

From this you will need to draw on a variety of fields to do this and this has a bit of variation depending on what kind of model you are dealing with, and what kinds of patterns you are trying to codify.

For example if you are trying to use these applications in the area of signals intelligence, then you will probably be incorporating things like Fourier transforms in your work. In other areas you might need to use different techniques to get the information you want because the transform required is the one that decomposes your data into the stuff that you actually need.

In addition to the above, it's important to know domain specific information in the same way that any other field has very narrow and specific domain knowledge.