MSc Mathematical Modelling and Self-Learning Systems
Self-learning systems are an important and newly emerging technique in many areas of applied science such as Applied Mathematics, Engineering, Computer Science and Statistics. In particular, self-learning systems are a disruptive approach to mathematical modelling which use differential equations at their foundation. A particular strength of this approach is that it combines numerical learning algorithms such as dynamic machine learning with differential equations to design applications that can adapt to a changing environment. This approach is new and unique because it explicitly takes into account the dynamic aspects of data and allows for fast and accurate modelling of self-learning systems. This is a new and rapidly developing area at the interface between applied mathematics and machine-learning (for example see here).
The primary aim of this course is to provide training in the use and development of modern numerical methods and self-learning software. Graduates will develop and apply new skills to real-world problems using mathematical ideas and techniques together with software tailored for complex networks and self-learning systems. While there is a strong focus on modern applications, graduates will gain in-demand skills in mathematical modelling, problem-solving, scientific computing, dynamic machine learning, complex networks and communication of mathematical ideas to a non-technical audience.
More general hands-on skills include mathematical typesetting, mathematical writing, desktop and web-based mathematical software development, and the use of computer languages and packages such as C#, R, Python and TensorFlow.
Part 1
**Students who have taken ST4060 or ST4061 in a previous degree must select alternative modules (subject to availability and timetabling) from list A and list B of fourth year of the BSc (Mathematical Sciences) in consultation with the Programme Coordinator.
Part 2
Self-learning systems are an important and newly emerging technique in many areas of applied science such as Applied Mathematics, Engineering, Computer Science and Statistics. In particular, self-learning systems are a disruptive approach to mathematical modelling which use differential equations at their foundation. A particular strength of this approach is that it combines numerical learning algorithms such as dynamic machine learning with differential equations to design applications that can adapt to a changing environment. This approach is new and unique because it explicitly takes into account the dynamic aspects of data and allows for fast and accurate modelling of self-learning systems. This is a new and rapidly developing area at the interface between applied mathematics and machine-learning (for example see here).
The primary aim of this course is to provide training in the use and development of modern numerical methods and self-learning software. Graduates will develop and apply new skills to real-world problems using mathematical ideas and techniques together with software tailored for complex networks and self-learning systems. While there is a strong focus on modern applications, graduates will gain in-demand skills in mathematical modelling, problem-solving, scientific computing, dynamic machine learning, complex networks and communication of mathematical ideas to a non-technical audience.
More general hands-on skills include mathematical typesetting, mathematical writing, desktop and web-based mathematical software development, and the use of computer languages and packages such as C#, R, Python and TensorFlow.
Part 1
**Students who have taken ST4060 or ST4061 in a previous degree must select alternative modules (subject to availability and timetabling) from list A and list B of fourth year of the BSc (Mathematical Sciences) in consultation with the Programme Coordinator.
Part 2
