Back to ListMachine Learning Theory
Level
Foundation, Broad, Theory, Methodological.
This topic covers the mathematical foundations of machine learning considering different learning paradigms and all facets of machine learning theory, including provable generalisation guarantees as well as optimisation and numerical analysis.
Learning outcomes
Content /Knowledge
Students should be able to:
- Achieve a solid background on the theoretical and mathematical aspects needed to analyse the theoretical performance limitations and possible improvements of modern machine learning algorithms under different modelling assumptions on the data and the learning agent.
- Understand computational and algorithmic aspects underlying different machine learning systems. To be able to understand and derive rigorous complexity guarantees.
Methodologicalskills
Students should be able to:
- Derive rigorous guarantees on the learning accuracy of different algorithms using a broad range of tools, e.g., from statistics, probability, game theory, optimisation, and stochastic processes.
- Derive rigorous guarantees on the computational efficiency of different algorithms using tools from fields including optimizsation, numerical analysis, and analysis of algorithms.
Transferrable/Application
Students should be able to:
- Work effectively with others in an interdisciplinary and/or international team.
- Design and manage individual projects.
- Clearly and succinctly communicate their ideas to technical audiences.