Introduction to machine learning
Deadline for application
6 September 2024
Machine learning methods currently represent some of the most powerful and dynamic developments in the financial sector. This course introduces the building blocks of machine learning and discusses selectedmethods, making connections between them and conventional statistical methods. The discussion of each method is followed by a practical session with examples and exercises in R.
This course also addresses the practical challenges associated with the adoption of machine learning. It provides a forum for central bankers, regulators and supervisors to present and discuss strategies to develop and implement machine learning models, thereby enabling an exchange of knowledge among countries on this increasingly important topic.
- Shrinkage methods
- Decision trees
- Ensemble methods
- Bias-variance trade-off
- Advantages and limitations of machine learning methods
- Discussion of case studies from course participants
Starting on the second day, there will be a Q & A slot before every session. Participants will have the opportunity to discuss the content of the previous day and any challenges they encountered with the practical exercises which involved independent programming in R.
The course is aimed at data-savvy central bankers, regulators and supervisors in areas such as information technology and statistics, or research departments interested in implementing machine learning methods.
Fundamental knowledge of data analysis (including linear and logistic regression) and statistical software (including commands in R) is required.
Participants are expected to make an active contribution to the discussions and will be invited to present and discuss current challenges related to the implementation of machine learning in their own jurisdiction. There will be opportunities to present independent analyses related to the topic of the course.
Computer with microphone, camera, speakers or headphones; an up-to-date internet browser.