fbpx

What is the Best Way to Learn Artificial Intelligence?

May 5. 2017. 8 mins read

In the fall of 2016, the citizens of the United States became fed up and made a decision that was heard around the world. While not all states felt the same way about this decision, it was pretty clear that it was time for some real change. The last 8 years had been an utter disappointment, full of some of the most mindless inane drivel the world has ever seen, making people truly question if the United States was really the great nation that it thought it was. Then, last fall, it finally happened.

On or around October 9th, the citizens of the United States of America collectively became more interested in artificial intelligence than Kim Kardashian’s net worth. After invading American homes for nearly 10 years, “Keeping up With the Kardashians” lost to AI with Google searches for “artificial intelligence” maintaining a sustained upward trend against Google searches for “Kim Kardashian net worth” which collapsed into a downward trend. While not all states felt the same about that, the results are pretty telling:

The blue states above decided that the era of AI has emerged, and we were quick to point out recently that this is not like the “dot-com era”. This time it’s different. If you’re pretty much anyone, one of your immediate concerns is probably the extent to which AI will affect the way you make a living. Did you know that being a truck driver is the most popular occupation in the United States? That’s hardly likely to last for more than a decade with autonomous trucks being developed as we type this. The world is changing and one way you can hedge your risk is to start learning more about artificial intelligence. This begs the question, what is the best way to learn artificial intelligence? The answer depends on who you are.

The Best Way to Learn AI for Investors

As an investor, you don’t need to be aware of all the technical details, but you do need to have a basic understanding of the concepts. The single most important takeaway we would impart is that anyone with a website can say they use AI and there is only one way to know that is true. Just like Elon Musk’s hair, the proof is in the outcome (this article helps that statement make sense). If we were to offer a quick Nanalyze course on “Artificial Intelligence 101″ right now, this is how our syllabus is going to look like:

Then, subscribe to our newsletter and we’ll keep you posted on investment opportunities in AI (there really aren’t any right now) and let you know when AI IPOs get filed (like this one). We’ll also warn you about companies that promise AI but haven’t quite delivered on that promise yet. In the dot-com era, retail investors invested in the promise of great things, but in the AI era, we invest in results only.

The Best Way to Learn AI for Young Adults

Nobody likes to hear this but if you choose underwater basket weaving as your major then you will not be gainfully employed afterwards no matter how much you get on Twitter and complain about how unfair the system is. We’ll put on our career adviser hats for a second and tell you this in the most delicate way possible. You absolutely need to study STEM or you will be fcuked. This means that for artificial intelligence, you’re best off learning how to sling code through majoring in computer science (though that isn’t the only road you can take to get there).

People with degrees in computer engineering, computer science, mathematics, and physics will be at a great advantage and will naturally levitate towards AI. Software development or programming is, in fact, a major building block of neural networks and deep learning and hence, essential to learning AI.  Mathematics, especially in algebra, calculus, algorithms, and statistics, are usually prerequisites for courses in programming, data science, neural networks, machine learning, and deep learning. Sorry, for those of you who hate math. You just need to bite the bullet on this one.

Browsing the webpages of the best universities out there shows that we currently don’t have a “B.S. in AI”. What we have mostly are degrees in Computer Science or Computer Engineering with a specialization in AI. Currently, there appears to be no exclusive AI curriculum except for the University of Georgia with its MS in Artificial Intelligence. What you can be absolutely sure of is that any halfway decent computer science department should be all over AI like a cheap suit.

Let’s aim high to start. The US News ranked the best graduate programs with specialization in artificial intelligence in the United States as follows:

  1. Stanford University – Stanford
  2. Carnegie Mellon University – Pittsburgh
  3. Massachusetts Institute of Technology – Cambridge
  4. University of California-Berkeley – Berkeley
  5. University of Washington – Seattle
  6. Georgia Institute of Technology – Atlanta
  7. University of Illinois-Urbana Champaign – Urbana
  8. University of Texas-Austin – Austin
  9. Cornell University – Ithaca
  10. University of California – Los Angeles
  11. University of Michigan-Ann Arbor –  Ann Arbor
  12. University of Massachusetts-Amherst – Amherst
  13. Columbia University – New York
  14. University of Pennsylvania – Philadelphia
  15. University of Southern California – Los Angeles
  16. California Institute of Technology – Pasadena
  17. University of Maryland-College Park – College Park
  18. University of Wisconsin-Madison – Madison

Sure, these places are tough to get into. Sure, the tuition is expensive. Sure, you don’t like math. But the truth is, this is where the future is heading. We helped you out by putting the link to each school’s website above so start filling out those applications. While we’ve listed the top universities out there, don’t be afraid if you get into a lesser university. Just study your azz off and then apply for grad school at one of the above universities when you’re done and you’re bound to get in one.

The nerds took over investment banking and they’re well on their way to taking over everything else. How bad do you really want it?

The Best Way to Learn AI for Working Professionals

As a working professional, you may be at the C-level trying to understand how AI impacts your business or you may be resting in the back of your long haul at the truck stop wondering what the future holds. In either case, you need to come up to speed with artificial intelligence because your livelihood is at stake. You don’t have time to get a degree because you have a family to support, so you want to get exposure to AI as quickly and efficiently as possible.

Artificial Intelligence Courses at Udacity

The best approach is to start with free introductory courses to artificial intelligence. How do we know these courses are good? Because the Chairman and co-founder of Udacity is this guy:

Learn Artificial Intelligence from Sebastian Thrun

While Sebastian Thrun doesn’t quite have the notoriety of global thought leaders like Rob Kardashian, he’s still someone worth knowing about. You can pretty much be assured that the long list of courses on artificial intelligence at Udacity are going to be solid because Dr. Thrun teaches a lot of them. Some courses have restricted attendance that you need to apply to be on a waiting list but you can start with the below free introductory courses:

After completing these free introductory courses in AI, you can then deep dive into more specialized areas through their Nanodegree Program. Here’s a sampling of these “nanodegrees”, all of which Dr. Thrun helps teach:

You can be pretty certain that companies are going to come around sniffing for AI talent once you’ve got that qualification. If they don’t, consider a boot camp.

Attend a BootCamp

A boot camp is a compressed and immersive version of a specialized course but emphasizes more application and problem-solving. With AI boot camps, it usually assumes you are already good at mathematics, have done some programming/coding, and have a basic knowledge of machine learning or deep learning. Here are some sample boot camps:

  1. Artificial Intelligence Immersive Bootcamp by Experfy
  2. NVIDIA Deep Learning Institute Bootcamp
  3. Dev Bootcamp: Artificial Intelligence 101, Data and Machine Learning with x.ai

The advantage of a good quality boot camp is the network of relationships standing behind the organizer. They usually have corporate partners already waiting to hire graduates or mining for talent they can take to one of the +1,500 AI startups out there. The cost of these boot camps may be a bit high and they may not be scheduled regularly.

Books and Forums

If your goal is to immerse yourself in AI without getting into any of the technical details, then you could be a C-level executive who just needs to make sure your firm is “using AI”. Here is a book you can leave lying around the office to make sure everyone knows you’re “getting it done”:

Casually mention to your colleagues that Elon Musk said this is the only comprehensive book out there about deep learning and machine learning. That should make you feel somewhat better about the $72 price tag. Finally, you can frequent forums such as Quora or Reddit to stay up to speed on the news.

Quora – Artificial Intelligence

In this forum, you can interact directly with Ben Hammer of Kaggle, Ian Goodfellow of Google, Yann LeCun of Facebook, Xavier Amatriain Quora’s VP of Engineering, and many others. It has 554,800 followers.

Reddit – Artificial Intelligence

Conclusion

While this article presents a summary of ways to learn artificial intelligence, we have some of our on-staff MBAs working diligently to put together “How to Learn AI – The Ultimate Guide” that we’ll make available to our lovely readers for free when it’s complete. While our guide won’t be winning the Teen Choice Awards anytime soon, it may just help make sure the teens of today are gainfully employed tomorrow.

Share

Leave a Reply

Your email address will not be published.

  1. Scikit-learn, the free software machine learning library for the Python programming language is the best way to study artificial intelligence, Anaconda, the freemium open source distribution of the Python and R programming languages for large-scale data processing, will install everything you need.

    But I think acquiring working knowledge in this subject will be extremely time consuming and therefore an expensive proposition. I’m actually a bit doubtful that I can master it well enough to do any original work. But you can certainly improve your understanding of the technology and it meshes well with my interest in cognition and the nature of consciousness.

    1. Thank you for the useful feedback Robert. Regarding the comment about how time consuming it is to learn and expensive. This means that for someone who has a day job that they would have to work loads after work? Is time the main expense here? For someone studying it in Uni, time should not be an issue then.

      1. Yes, your time would be the major expense. You should also consider getting an expensive NVIDIA GPU-accelerated graphics card for processing power, but I think you can just lease time on the cloud. There is a lot of advanced math involved and I doubt that I could learn that without years and years of effort.