Title:	Diffusion models for data-driven risk management
DNr:	Berzelius-2025-402
Project Type:	LiU Berzelius
Principal Investigator:	Joakim Andén-Pantera <janden@kth.se>
Affiliation:	Kungliga Tekniska högskolan
Duration:	2026-01-16 – 2026-08-01
Classification:	10210
Keywords:

Abstract

The success of deep learning methods rests largely on their flexibility in adapting to complex data, but they often amount to a black box and require a large amount of data. As a result, pushing research in financial analytics towards explainable and robust learning methods is essential for the industry to adopt deep learning on a large scale. The goal of this project is to develop and analyze explainable deep learning methods for applications to financial risk management. Previous work has identified promising models for this end, but efficient sampling remains a challenge. We will explore this topic though the lens of diffusion models, aiming at accelerated sampling. These models will also be evaluated against classical MCMC models. Additional topics in the application of diffusion models, such as constrained sampling, will be studied.