What is Machine Learning?

Learn what distinguishes machine learning from traditional rule-based programming, how machine learning systems learn from data and improve over time, and how this approach fits within the broader field of artificial intelligence as the dominant method for tasks requiring adaptability and pattern recognition.

Understand how the California Housing dataset represents real-world information as tabular data with rows as block groups and columns as numeric features, recognise different variable types, and see how visualising distributions with histograms reveals patterns in the data.

See how a machine learning model can be trained to predict the median house value in Californian districts based on input features, and how its predictive accuracy is assessed using new data it has not seen before.

Understand the fundamental structure of a supervised machine learning problem by identifying the roles of data (inputs and labels), the model and learning algorithm, and the distinction between the training and inference phases.

Understand the main types of machine learning tasks by distinguishing between supervised, unsupervised, semi-supervised and reinforcement learning, and learn how supervised problems can be classified further as classification or regression based on the nature of the output.

Understand how to analyse real-world problems to determine whether machine learning is appropriate, frame problems as classification or regression based on the desired output, and identify suitable inputs, outputs, and learning setups for effective model design.

Understand why the primary aim of supervised learning is to build models that generalise well to unseen data, and how test sets are used to estimate generalisation error as opposed to training error.