I'm a first year university student and I want to learn about AI and machine learning specifically. Where should I start? What sorts of resources should I use?
Ask HN: Getting Started With Machine Learning/AI
1–5 of 5 posts
Re: Ask HN: Getting Started With Machine Learning/AI
#2I just started watching some Machine Learning lectures on iTunes U. from Stanford. Great stuff so far.
http://deimos3.apple.com/WebObjects/Core.woa/Browse/itunes.s...
Re: Ask HN: Getting Started With Machine Learning/AI
#3The good news is you are at the right place :)
You should probably pick up courses on machine learning as will I.
http://www.cc.gatech.edu/classes/AY2009/cs7641_spring/
kind of tells you the topics you may want to look at.
http://www.autonlab.org/tutorials/
is mentioned in the previous threads which has some neat tutorials too
Re: Ask HN: Getting Started With Machine Learning/AI
#4I know this is obvious, but... Find out which of your professors work in this area. Then go talk to them, and ask them about their work. Get involved with a research project.
Learning things from books is great (and you'll still have to do a lot of that), but there is no substitute for networking. Approaching professors can be intimidating, but you'll have to do it if you want to go to grad school. Might as well start now.
In my experience, the profs who are really good at what they do don't act arrogant and aren't dismissive of students. If anything, they get really excited talking about their research, and you'll have a hard time leaving their office :) If you do encounter someone who is arrogant, that tells you something about them (you don't want to be their grad student). However, don't mistake a busy prof for an arrogant one. Sometimes you just have to be persistent.
Re: Ask HN: Getting Started With Machine Learning/AI
#5A good indespensible textbook for me had been chris bishops book Pattern Recognition and Machine Learning. However it might be suited best for the grad student.
Honestly the best thing you can do is start working on machine learning problems. Try and build a classifier or predictor. Brush up on probability theory.
If you have no idea where to begin, check out Collective Intelligence. It's a great intro read that gives a good background into the types of problems ML and AI are trying to solve.