Machine learning for the classroom

Machine learning is a hot topic. It is a subset of artificial intelligence and is about teaching a computer, rather than programming it. One of the approaches in machine learning are artificial neural networks (ANNs) that mimic the function of our brains. Though much simpler than their biological counterparts, they are impressive in what they can do, especially when it comes to image processing.

I recently had a conversation with my year 8 twins about learning strategies. I tried to explain to them how their brains learn and why, therefore, breaks are important to give the brain time to 'digest' new information (aka to consolidate the molecular updates that regulate the exchange of information between the neurons). So I drew a couple of neurons on a sheet of paper, interconnected them and showed how a brain learns and thinks.

A neuron (artist impression).

I thought I could do better than this. I wanted my boys to look inside and experiment with an actual neural network. If you followed some of my earlier work with the B4 computer processor kit, you would have guessed that I am the kind of person that likes to open windows into things that are otherwise hidden. I have previous experiences with ANNs which gave me a head start. I tweaked an ANN that I made previously and created the UI to look inside.

Here is a screenshot of an experiment where the ANN has been trained and is able to recognise the letters A .. J.

A trained ANN recognises the letter B.

On the left, the eye sees a pattern in a 5x7 grid. It forwards the bits via the optical nerve to the brain that consists of three layers (input/hidden/output). On the right, a nerve cell (perceptron) fires. The network has recognised the B.

We can play with the information that the eye sees. Clicking into the grid, we can set and unset pixels. As we gradually modify the B, the ANN will be less sure about this being a B. Is it perhaps a D?

B or a D? Bit of both!

Changing the pattern one bit further towards the D, the ANN gets much more confident that this is a D.

More B than D.

So neural networks are great at dealing with information that is not quite ideal. This is often the case in the real world, especially when it comes to image processing. Imagine a self-driving car. The ANN produces a probability value, the more black a box, the more certain it is.

From the perspective of an ANN, anything is a pattern. So we let it loose on a couple of emojis:

Our ANN recognises emojis.

But what about more complex patterns? Could our simple ANN convert decimal numbers into binary? Yes, it can:

ANN recognising the number 5 and converting it into binary.

This educational ANN has three special features:

1. The learning process is short, animated and is entertaining to look at.

2. The line colours change subtly as the network learns. You can observe the network learning and the learning is in the network. Each network has its own fingerprint.

3. You can observe the perceptrons (nerve cells) firing as the network processes information.

The network has been programmed with different learning strategies. It can also forget information. Your students will see why revision is important and understand why they sometimes forget things they have learned (see experiments 6 and 7).

The website mycomputerbrain.net hosts the ANN. Your primary and secondary students can experiment with it. I have prepared a couple of experiments (with instructions suitable for school kids), and more might follow. Any ideas?

I think the ANN would be of interest to teachers of Digital Technologies or Biology, but generally to anyone who teachers growth mindset and learning techniques.

Enjoy. Let me know how you go via the contact form, or just on Twitter. Twitter @DigTecInstitute #ai

Karsten

B4 in the Classroom - Teacher Feedback

​I have been teaching the WACE Computer Science course for many years, and each year I have struggled to come up with a way to teach the Systems Architecture component that is engaging for the students. Yes, I have lots of old components like motherboards, RAM chips etc. that the students can touch and see, but ultimately when it comes to them understanding processes like the Fetch-Execute cycle and how the CPU interacts with RAM, the best I have been able to do is find some animations on the internet to make it more interesting.

Well, that was before I found the B4’s! I first encountered them at the 2017 ECAWA Conference and I was absolutely blown away not only at the simplicity of the design, but by the genius of how it brings alive to the students’ concepts that were dull and boring before.

​For the Year 11 students, the B4 makes learning Binary numbers a breeze. For the Year 12 students, they now GET what really happens inside their computer, because they can see it happening in front of them!

I started the lesson with instructions about connecting the different wires to the modules, then stepped the students through the first couple of experiments from the handbook. Once they had gained confidence in using the B4’s, I let them loose to experiment by themselves. I have never had such a quiet classroom! – the students were all fully engaged in what they were doing and didn’t want to stop when the lesson time ended.

For the next lesson I got the students to read back through the Systems Architecture notes, we discussed different things that the students encountered when they were experimenting with the B4’s and for the few things that they still were unsure of, we got the B4 kits out again and went through the concepts again.

Feedback from the students has been very positive – they love the fact that they now have something concrete to hang their learning on. We are about to start the Programming unit of the course, and I can’t wait to incorporate the B4’s into my programming lessons.

We are walking supercomputers

Let's assume you gave one of your skin cells to a forensic scientists and asked them to extract and uncoil its DNA. How long would this string of DNA be? Estimates vary between 2 and 3m. Let's go conservatively with 2m for what follows:

Source: Shutterstock

Most of our body cells contain a full copy of our DNA. How many cells are in a human body? Nobody has counted them yet, but estimates are in the 37 trillion range. That's 37,000 billion cells - a 37 with twelve zeros. In full: 37,000,000,000,000.

37 trillion cells each having 2m of DNA, that's 37 trillion x 2 metres = 74 trillion metres of DNA in just one human body.

How much are 74 trillion metres? Divide by 1,000 and we are dealing with a more comfortable scale of kilometres: 74 billion km. That's still a lot. Let's put this into perspective.

• From Perth to Brisbane, we travel some 4,000 km.

• Around the globe: 40,000km.

• The distance between the Earth and its moon is 384,000km.

• To the sun? 150 million kilometres.

We have to go further to the edge of our solar system.

The distance between Earth to the dwarf planet Pluto is 7.5 billion kilometres, a mere 10% of the combined DNA of one human being. It took NASA's New Horizons Spacecraft 9 years, 5 months and 25 days to get there. But we are making progress in finding a suitable scale to visualise the length of our DNA chain.

(c) NASA

So, the equivalent distance of the length of DNA of just one person is 5 return trips from Earth to Pluto.

The Voyager 1 spacecraft, launched in 1977, is now the furthest space probe from Earth at a distance of 21 billion km. It has travelled just over ¼ of the length of our DNA chain. You can track its progress here at NASA's Jet Propulsion Laboratory website at Caltech.

What about humankind?

Well, let's say we have (rounded) eight billion people on this planet.

We multiply the world's population by the length of one person's DNA:

8 billion people x 74 billion km = 592 billion billion km.

How much are 592 billion billion km? We are now truly entering astronomical scales well beyond our solar system. Let's express this distance therefore in light years. One light-second is the time it takes light to travel about 300,000km in the vacuum of space. And because one year is 60 sec x 60 x 24 x 365 = 31,530,000 seconds, therefore 1 light-year is equivalent to 9,461,800,000,000 km. (9.5 trillion km)

We divide our combined human DNA length of 592 billion billion km through 9,460,800,000,000 km and the result is 62,573,990 light-years.

Let's round this to 63 million light-years so that we can work with this number more easily.

The diameter of our galaxy, the Milky Way, is 100,000 light-years. 63 million light-years of human DNA divided by 100,000 light years is 630 trips across the Milky Way, or 315 return trips.

Source: Wix

This is only the code that runs in the human species. We haven't even looked at the DNA of plant and animal life. We are talking about several orders of magnitude.

Let's take this to the cosmic extreme: The diameter of the universe is estimated to be 46.5 billion light-years, which is 740 times the length of the combined human DNA chain. If we included the DNA of all plant and animal life on Earth we should be able to cross the universe with this DNA chain or at least get close to crossing it.

Supercomputers

Our cells are processing the information from their respective DNA molecule on an ongoing basis in a massively-parallel effort. They make proteins, regulate cellular activities and communicate with other cells. This makes each cell a computer with a built-in sophisticated 3D protein printer (I am simplifying) that can manipulate matter at the individual electron and proton level. While we usually only consider the brain (nerve cells) when talking about the human-computer analogon, we forget the underlying computing tasks carried out by the rest of our body cells.

So yes, we are supercomputers.