Deep Learning for Actuarial Modeling

About

This summer school aims to give an overview and an introduction to the latest developments in the field of actuarial modeling.
We start by providing a solid technical basis on statistical modeling by introducing the familiar framework of generalized linear models (GLMs), which is based on the exponential dispersion family (EDF) of distributions.
This family encompasses the most important distributions for regression modeling in actuarial science, including the Poisson model, the gamma model, and Tweedie's model.
With instructors from renowned academics and global experts in the field, the course allows Access to exclusive resources, such as course materials, research papers, and codes

Target group

This program is designed for actuaries, data scientists, and professionals in related fields who are interested in enhancing their skills in deep learning and actuarial modeling.
The participants should have basic knowledge in probability theory, statistics and actuarial modeling. They should be familiar with R and/or Python. Practical experience is an advantage, but not necessary

Course aim

Exposure to cutting-edge deep learning techniques specifically tailored to actuarial modeling;
Detailed case studies demonstrating the practical application of advanced methods in actuarial practice
The methods that form the main tool to bring covariate information into a suitable tensor structure for more advanced deep learning tools. These are then presented, like the recently developed attention layers and transformers, which are the core deep learning modules in large language models (LLMs) such as ChatGPT. We also discuss the credibility transformer, which integrates Bühlmann credibility into the transformer architecture. This credibility mechanism improves model fitting and it integrates explainable features into the transformer architecture.
This summer school is completed by discussing several special deep learning architectures, such as the LocalGLMnet, which aims at mimicking the behaviour of a GLM, in-context learning (ICL), foundation models or negative sampling.
This school is supported by hands-on exercises in R and Python. We have prepared many different notebooks that illustrate different aspects of our methodological presentations.

Fee info