Does learning AI need intelligence? Yes, of course. The intelligence to recognise that AI is becoming more and more important in every aspect of business and science. Does it need a formal education? No, just intelligence. But how to go about it?

Old and worn shoes

Should we just go on using our old shoes, or go through the pain of adjusting our feet to new footwear?

It may be tempting to say that what was good enough for me then, is still good enough for me now. But this evades the issue: AI is getting more prevalent every day, and we need to adapt to the new reality. What is the best way to accomplish this? As painlessly as possible, of course.

For the sake of argument, let’s assume that the kind reader is not currently studying (at university, or other similar institution) but is intelligent and willing to learn, to spend some time getting their feet wet in the newfangled jungle of AI.

There are a number of steps to be taken; we cannot fly before we master crawling, as well as walking and running. One step at a time.

Science and AI - must they always meet? The answer is no, it is perfectly possible to understand and use AI without a scientific background. What then do you really need? An understanding of what you want - what is your goal.
Blackboard with scientific drawings

Depending on your level, I would suggest you start with the elements of AI This is an excellent introduction to the basics of AI, the necessary jargon, and what it accomplishes. As the introduction to the course says,

Are you wondering how AI might affect your job or your life? Do you want to learn more about what AI really means — and how it’s created? Do you want to understand how AI will develop and affect us in the coming years? Then the following holds: Our goal is to demystify AI.

The Elements of AI is a series of free online courses created by Reaktor and the University of Helsinki. We want to encourage as broad a group of people as possible to learn what AI is, what can (and can’t) be done with AI, and how to start creating AI methods. The courses combine theory with practical exercises and can be completed at your own pace.

The course is meant for beginners, those curious what all the buzz is about. The first 6 chapters are

  1. What is AI?
  2. AI problem solving
  3. Real world AI
  4. Machine learning
  5. Neural networks
  6. Implications
This is just the first part (Introduction to AI); it will be followed by a second part later in 2020 (Building AI).

Note that this course is rare in that it is also available in other languages but English: today in Estonian, Finnish, Latvian, German, Swedish and Norwegian. The stated aim is to have it in all official languages of the EU. Not bad.

But there are lots and lots of resources on the web for AI. For example, have a look at Google. Google is one of the foremost users as well as resources for AI on the web today. Their excellent https://ai.google/education/ gives you virtually everything you want, especially if you are a practitioner: just choose your level and off you go:

AI profile selection
Target filter in ai.google. Other choices are content types (Courses, Videos, Interactive, etc.), or stages of ML development (Data, Preparation, Developing, Deployment, etc.).

Google is also one of the biggest contributors to open source software in the AI sphere.

But Google is of course not alone. Have a look at EdX for on-line courses in AI. They have a very nice collection of education material, sourced from Harvard, IBM, University of California at San Diego, among others.

If you want something more academic virtually all the big universities offer on-line courses in a variety of topics. For instance have a look at https://www.deeplearning.ai for a collection of excellent resources, not all of them academic. Not sure at which level you are? Take their test and find out! (https://workera.ai).

The best way of learning something is almost always by doing: learn to code AI. There the premier (but not only one) computer language is Python. It’s available for all major platforms, like Microsoft Windows, Apple MacOS, and Linux. Apple and Linux has it even built-in. But it’s just a language, so you need other resources as well. For the more scientifically minded of us I recommend the excellent Scikit-Learn. They cover all the bases:

  • Classification
  • Regression
  • Clustering
  • Dimensionality reduction
  • Model selection
  • Preprocessing

And with excellent examples as well. You cannot go wrong with these resources.

So there you have it. Learning is easy, but of course motivation is everything. But if there is a will, there is a way.