The lecture introduces several fundamental concepts from machine learning and treats important  applications  in finance. It will cover topics like

-Neural networks

-Universal approximation theorems,

-Stochastic gradient descent,

-Backpropagation.

The financial applications include

-deep hedging,

-deep portfolio optimization,

-deep simulation and

-deep calibration.
 

After completing this class the student will have the ability to...

-theoretically understand  neural networks, stochastic gradient descent, reservoir computing, etc.

-apply modern machine learning techniques to solve problems arising in quantitative finance, like hedging, portfolio optimization, prediction and calibration tasks.

Standard rules for PIs

This class is taught as a lecture complemented with exercises and a coding project.

Exercise Series (25%)

Coding project (20%)

Final exam (55%)

• Successful completion of Mathematics I and Financial Markets and Instruments
• Successful completion of at least 42 ECTS credits from the first year compulsory common courses
• Allocation to the elective