Important questions remain unanswered because the data necessary to address them is encoded into high-dimensional data structures such as images or language. Economists have become increasingly interested in using machine learning models to transform these data into simpler representations, and use them for economic analysis. After introducing predictive modelling, this course provides a comprehensive understanding of neural networks models, their optimisation, as well as specialised model architectures to make sense of these complex forms of data. From a strong base in theory and mathematical formalisation, focus is kept on intuition and efficient implementation using Python.
