Artificial Intelligence: Learning from Data

Artificial intelligence and machine learning are appearing everywhere. Our previously “dumb” machines are becoming “smart” and the organisations we deal with on a daily basis are increasingly using AI to make decisions about us.

But don’t let the workings of AI be a mystery to you! Discover how algorithms and data come together to create the illusion of intelligence!

In this second of three AI courses, you will learn how to use data to train an AI to solve problems. Given a maze or map, we will use the reinforcement learning technique to get an AI agent to navigate the best route through the map. We will then apply the same technique to get a robot to learn to walk!

This course can be taken individually or as part of the series of three.

Course Outline



Suitable For



Gain hands-on experience in using data to build AI to make predictions and decisions with the Python programming language.


Session 1: Agents

We will introduce the idea of a software agent that learns by example.

Session 2: Maze Solving

We will build a computer-based representation of a simple maze and use a machine learning approach called reinforcement learning to teach our agent to solve the maze.

Session 3: Pick-up and Delivery

We will use the techniques we learned and apply them to the more complex task of picking up items from one location and delivering them to another.

Session 4: Teaching a Robot to Walk

We will extend the techniques to teach a robot to move.


Prices depend on location. See individual dates for details.

Skills you should have before you attend

Those attending the course would benefit from having a good solid understanding of programming, preferably using Python.

Skills you will pick up in the course

Key skill categories

On successful completion of this course, students will be able to


This course can be customised and delivered at a mutually-convenient time and place for groups


Notify me when new dates become available.