VVZ API is not affiliated with ETH Zurich. Data might be outdated or incorrect. Please view the official ETHZ Vorlesungsverzeichnis for binding information.

401-3931-00L 4 Credits BSC , DR , MSC D-MATH
You're viewing possible stale or outdated data. Please check the latest semester for more up-to-date information.

Responsible Machine Learning with Insurance Applications

Lecturers & Examiners: Michael Mayer, Christian Lorentzen-Geiser
VVZ CR n/a

Last Updated: 2026-02-05 16:02:14

Abstract

This lecture covers important aspects of applying supervised machine learning models in a responsible way, based on sound statistical theory. The focus is on model interpretability, calibration (bias) assessment, and proper model comparison. The methods are illustrated with actuarial datasets.

Objective

The student is familiar with the main tools of model interpretability, calibration assessment, and model comparison and knows how to apply supervised machine learning in a responsible way.

Content

• Overview of supervised machine learning (statistical learning theory, GLMs, tree based methods, and neural nets; cross-validation) • Model interpretability methods (partial dependence plots, measures of variable importance, and SHAP) • Bias/calibration assessment with identification functions • Model comparison with consistent scoring functions • Working with dependent observations and further topics

Resources

Learning Materials (Links)

General Information

Language
English
Levels
BSC , DR , MSC
Frequency
Yearly recurring

Examination

Type
session examination
Mode
oral 30 minutes

Course Components

Type Title Time & Place Hours
lecture with exercise Responsible Machine Learning with Insurance Applications
  • Mon 14:15-16:00 (CAB G 56)
2 h weekly

Offered In