Introduction to AI

This course is aimed at the Python programmer who wants to learn how to apply deep learning to data problems.

In this course we will learn about how a neural network is structured with multiple hidden layers and non-linear activation functions such a way that we can input some features of data to try and predict some labels.

For the purpose of this course we will be using a tool called Jupyter Notebooks which provides you with a local editor and Python terminal in your web browser.

How to use the course materials

You can follow the course materials online.

To run the code on your own machine, follow the setup instructions.

You can also access individual pages from the course as Juypter notebooks.

Audience level

You should be comfortable programming in Python. Experience with the content of our Applied Data Analysis in Python course or equivalent is strongly recommended. This covers:

  • Running python in a Jupyter notebook
  • Working with dataframes/numpy arrays
  • Plotting with matplotlib/seaborn
  • Defining functions
  • Using classes such as sklearn models
  • Test train split and validating models

The theory of deep learning includes ideas from calculus and linear algebra. Understanding the idea of a derivative of a function would be strongly recommended. Having some notion of matrix multiplication would be useful but is not essential.

Intended learning outcomes

By the end of this course, you will:

  • Understand the basic theory of a feed forward multi layer perceptron.
  • Start to get the grips with PyTorch, tensors and writing classes in Python.
  • Understand how to pre-process data for training (including test train split)
  • Understand how to feed forward data and evaluate test loss for a neural network.
  • Have a basic intuition for what gradient descent and back propagation are.
  • Implement back propagation to update our weights and biases and reduce our test loss.

What this course will not cover

Deep learning and neural networks are a huge field of active research that we cannot cover in 3 hours. This is an intro level class designed to cover the learning outcomes above and serve as a prerequisite to further topics in AI, neural networks and deep learning, such as:

  • Cross entropy loss and more advanced optimisers
  • Training neural networks using High Powered Computing (HPC) resources
  • Convolutional Neural Networks (CNNs) for image/video analysis
  • Recurrent Neural Networks (RNNs) for time series and natural language processing
  • Transformers and Large Language Models (LLMs)
  • Graphical Neural Networks (GNNs)
  • Reinforcement Learning (RL)