A Parent’s Guide to Machine Learning for Kids and Teens
Artificial intelligence and machine learning have become such buzzwords today - most of us have heard these terms often but most of us really don’t know what they really mean or how it all works.
As parents, we have an excuse - we didn’t grow up with it and it wasn’t really around. It’s a more recent trend. I must add that this trend is not going away, only going bigger and wider, encompassing every field around the world.
For our children, they do not have a choice - they must understand how this magic happens, just like they need to understand how day and night is caused by earth’s rotation or what photosynthesis means.
In this guide, I’ll help you understand these terms - artificial intelligence and machine learning, discuss the benefits of introducing machine learning to our kids, at what age should we introduce our kids and teens to machine learning and, most importantly, how to best do it.
There are 3 sections in this guide.
Should our kids learn how to build machine learning applications
The best way to introduce machine learning to kids and teens
What is Artificial Intelligence
Artificial intelligence (or simply AI) refers to tasks done by a computer, tasks that would normally need human intelligence. There are various techniques used to mimic human intelligence. Some of these techniques do not even involve any learning on part of the computer.
For example, in 1997, a computer called Deep Blue defeated chess world champion, Garry Kasparov at chess. It was an AI system, but it was just programmed with a set of rules and strategies to win. Deep Blue simply followed the instructions coded into it, step by step.
It was, in no way, smarter than any human, let alone chess grandmaster, Garry Kasparov. But it won because it was able to calculate potential moves quicker and more accurately than he could, simply by following instructions that humans programmed it to.
On the other hand, there are AI techniques that involve “learning” from past examples. This leads us nicely into machine learning, one of the techniques very widely used today for artificial intelligence.
What is Machine Learning?
Traditionally, a computer does something by following step-by-step instructions that we program it with. Machine learning, on the other hand, performs a task by learning from various examples.
Here is a classic example. Computers now are very good at identifying cats in random images, something that humans do very easily but used to be extremely hard for computers. Using machine learning, a computer can be “trained” by exposing it to various images that have cats and various others that don’t have cats. Based on this “test” data, when a new image is presented, a computer can make a (fairly accurate) judgement of whether the image has cats or not.
Some Examples of Machine Learning Applications
Movie and show recommendations on Netflix are based on machine learning. Based on your viewing patterns and others’ who have similar viewing patterns, it makes predictions about what you are most likely to watch.
Voice recognition and understanding language such as when you use Alexa or Siri is also learned.
Here’s another classic example - an email spam identifier application will look at various examples of spam emails and not-spam emails and based on this experience, be able to identify a new incoming email as spam or not.
Should kids really be learning how to build machine learning applications?
Here is my personal experience - I tried to learn machine learning by enrolling into online courses - I found it hard and very time-consuming, especially if you want a deep understanding.
Of course, you do not need a deep understanding if you simply want to learn about potential applications of machine learning. And if you are simply interested in building simple machine learning based applications, you might get away with not needing to learn the nuts and bolts.
But you will still need to be fairly proficient in coding.
The point here is that for kids, their time is better spent in learning how to code first, before delving into AI and machine learning.
Having said that, done in the right order, there are many good reasons for introducing kids to machine learning and artificial intelligence, as you will see below.
At a fundamental level, all they will learn in machine learning courses offered by so-called kids coding education companies is some tools that use machine learning in the background. So, really you are dragging and dropping stuff and ‘coding up’ machine learning applications. You will not learn how machine learning actually works. You will not learn how to build machine learning models either.
But if you do hit a jackpot and enrol into one of the better courses, it can give you a flavour of the process of building applications that use AI, which brings me to the next section.
Benefits of introducing kids and teens to machine learning
Career prospects or expanding your child’s data fluency - whatever that means - are not reasons for introducing machine learning to kids. Here are some good reasons.
Exposing them to a variety of applications and the capabilities of machine learning will surely help fire their imagination of what might be possible and, I hope, kick off a lifetime of innovation, creativity and invention.
For our inventor / maker / techie kids, machine learning applications will get them started in their journey to build apps and systems that will make our lives easier and help solve the hard, global problems using tech.
At Riva Learning, we are highly focussed on teaching our students how to solve problems using technology. Machine learning is a very different way of problem solving and enables solving problems that seemed impossible for machines before. And because machine learning is so ubiquitous, kids need to have it in their arsenal to solve problems using this way.
Machine learning is all around us and is increasingly becoming a part of every system around us in every walk of life. It is not good enough for our children to just be consumers of this tech. They need to understand how it works, what the implications are and how it can make our lives better. Kids learn about planets and laws of motion not to become astronomers but because it helps their understanding of the world around them.
Here is a little example of why kids should understand machine learning.
Law and policy makers need to understand machine learning to make sure appropriate protections can be put in place while supporting innovation and development. Policy makers include hospital boards, company directors, charity trustees, school governors - people making the decisions - and they are ones who surely must understand how it can and should be used if it’s going to be facing us every minute of our lives.
Finally, the good news is that it can be great fun experimenting with machine learning, just in case fun is a good starting reason to learn about AI and ML.
What is the Right Age for a Machine Learning Course for Kids?
Well, if you decide for your child to do a machine learning course anyway, I’d say no sooner than 11 years of age. This is based on what I see everyday among hundreds of our students.
Before that, they likely do not have enough context or motivation to take such learning forward. Of course, this is a generalisation, but surely 7 years - as some websites tout - is too young. My son is almost 7 and I can’t imagine formally teaching him about machine learning.
How to introduce machine learning to kids?
The goal should be for a child to have a solid understanding of
the capabilities of ML and
the process that goes behind building ML/AI applications.
That will help expand their imagination and motivate them to build their own applications - ones that solve their little or big problems and make lives easier.
The first step in this process has to be for kids to learn how to code - even if it is just being comfortable with building Scratch games and stories. They don’t have to be Pythonistas or expert Python / Java / C ++ programmers, even though it will help. (As a shameless plug, I would love for your child to join one of our holiday camps or weekly classes).
And if they are at least 11 years old at this point, that might be a great time to introduce them to machine learning and artificial intelligence. Having said that, at any point after is totally fine too. For instance, 15 years is just as amazing. Unless you are a techie yourself, you might find it hard to help them with it at home. And you might have to get some expert help, such as our upcoming course in machine learning and artificial intelligence.
The best way for kids to learn about machine learning is for them to get hands-on, first-hand experiences.
If they can train a computer to do something, they’ll get it.
If they see what makes it learn more effectively, or what sort of training makes it get everything wrong – they’ll get it.
If presented in the right way, the basic principles are totally accessible.
And for them to get experience with all of this, there are some great online resources as below.
The best resources for machine learning for kids
By far the most sensible resource that I have found in this field is Dale Lane’s https://machinelearningforkids.co.uk.
“This free tool introduces machine learning by providing hands-on experiences for training machine learning systems and building things with them.
It provides an easy-to-use guided environment for training machine learning models to recognise text, numbers, images, or sounds.”
The tool has a number of examples that you follow to build some amazing applications and of course, create away your own ones.
Of course, we’d love for your child to join our upcoming machine learning course. Register your interest here.
Credits
I have to give credit to Dale Lane for this article because I have adapted his thoughts with my own experiences to write this article.
Gobind Bansal
Gobind is the founder and CEO of Riva Learning. Computer scientist by training, he has a Bachelors degree in Computer Science and Engineering from the Indian Institute of Technology (IIT), a dual Masters degree in Computer Science from Massachusetts Institute of Technology (MIT) and National University of Singapore (NUS), and an MBA from London Business School. He has worked in various technical and business leadership roles with Amazon, Google, IBM and M&S.