Humans can often perform a task extremely well (e.g., telling cats from dogs) but are unable to understand and describe the decision process followed. Without this explicit knowledge, we cannot write computer programs that can be used by machines to perform the same task. “Machine learning” is the study and application of methods to learn such algorithms automatically from sets of examples, just like babies can learn to tell cats from dogs simply by being shown examples of dogs and cats by their parents. Machine learning has proven particularly suited to cases such as optical character recognition, dictation software, language translators, fraud detection in financial transactions, and many others.

The aim of this module is to provide students with an understanding of the role of artificial neural networks (ANNs) in computer science and artificial intelligence. This will allow the student to build computers and intelligent machines which are able to have an artificial brain which will allow them to learn and adapt in a human like fashion.

This module is a mix of theory and practice with big data cases in finance. Algorithmic and data science theories will be introduced and followed by a thorough introduction of data-driven algorithms for structures and unstructured data. Modern machine learning and data mining algorithms will be introduced with particular case studies on financial industry.

Equip yourself with principles of allocation and mechanism design from an operational perspective. Auction design and market microstructure of the stock market, liquidity provision in electronic financial markets such as dark pools, and capital adequacy of centralized clearing platforms are some of the specific applications that will be studied in the first part of this module. During the second part, you will be introduced to complexity economics of self-organisation, network modules, and strategic proteanism. Finally, you'll use network models to study economic interactions.