Module 2 – Machine Learning


SLIDE 10 (Quick Draw): Quick Draw is a great example of a game built with machine learning. You draw, and a neural network tries to guess what you’re drawing. Of course, it doesn’t always work. But the more you play with it, the more it will learn.

BONUS (Pixel-It): The purpose of this activity is to demonstrate how a computer classifies images as well as how a computer reads them. This is a simplified example of how a machine might use training data to tell the difference between two different letters (in this case, the difference between a letter ‘A’ and a letter ‘M’).

Included below is ‘Pixel-It-Blank’ (An Excel spreadsheet to share with your students), Pixel-It-Instructions’ (A set of step-by-step instructions to accompany the activity), and ‘Pixel-It-Unplugged-Activity-Sheet’ (An alternative version of the activity which can be delivered without the use of computers):

BONUS (Teachable machine):
Interactive activity which demonstrates how Supervised Learning works.
Teach a machine using your camera or images from your computer, live in the browser. No coding required: Image Model – Teachable Machines

BONUS (Pix2Story):
Pix2Story is an experiment in teaching an AI system to be creative, be inspired by a picture and take it to another level. Pix2Story allows users to upload a picture and, using Natural Language Processing, get a machine-generated story based on a genre of your choice (Adventure, Sci-fi, Thriller etc.) Give it a try:


SLIDE 22: How machines learn.

SLIDE 42: Example of reinforced learning:

This video shows how the Deepmind AI learned to walk simply given the end goal to achieve and the reward mechanism. Why does the way the model walk differ from the way we do so?