Machine Learning & Generative AI: A Hands-On Guide to Actuarial Practice

Einleitung/Dauer

Organised by the EAA – European Actuarial Academy GmbH in cooperation with Deutsche Aktuarvereinigung e. V. (DAV).

Machine learning (ML) and generative artificial intelligence (GenAI) are among the most transformative technologies shaping the future of the insurance and financial industries. Actuaries and data scientists are increasingly expected not only to understand these technologies but also to apply them responsibly and effectively in their professional practice.

This two-day seminar offers a comprehensive journey through machine learning on the first day and generative AI on the second. Hands-on coding exercises are included throughout, primarily using provided code snippets for experimentation and modification.

The first day covers traditional and modern machine learning, beginning with the foundations of supervised learning. Participants will explore both well-established methods, such as generalized linear models (including standard linear regression), and modern machine learning models like CatBoost, artificial neural networks, and ensemble methods. Strengths and weaknesses of these models will be discussed, along with the crucial topic of interpretability and explainable AI. The programme also includes special machine learning topics – such as unsupervised learning, time series forecasting, synthetic data generation, and recent innovations – which may be selected and emphasized according to the interests and preferences of the participants, ensuring that the content is tailored to their needs and provides a broad, up-to-date view of the field.

The second day focuses on generative AI, beginning with the basics of working with large language models (LLMs), including prompting techniques, pitfalls to avoid, and practical exercises using APIs or local setups. Building on this foundation, we will explore advanced concepts such as function calling, fine-tuning, structured outputs, and retrieval-augmented generation. The seminar concludes with an introduction to the emerging paradigm of agentic AI, where participants will see how LLMs can act as autonomous agents in applications like automated reporting or legacy system migration.

By combining conceptual insights with guided, hands-on coding exercises, the seminar equips participants to responsibly, creatively, and effectively apply machine learning and generative AI in real-world actuarial practice.

The seminar will be held in person, giving participants the opportunity to learn on site alongside other actuarial professionals, exchange ideas directly, and receive immediate support from the lecturers. The venue is the 4-star Novotel Munich City in Munich (details below). The evening event on the first day is giving participants the opportunity to connect, discuss practical questions, and build their professional network in an informal setting.

Preliminary Programme
Monday, 8 June 2026 - Machine Learning
08.45 - 09.00 Registration
09.00 - 09.15 Introduction and Welcome (EAA)
09.15 - 10.45 Foundations and Traditional Machine Learning
10.45 - 11.00 Coffee Break
11.00 - 12.30 Modern Machine Learning Techniques
12.30 - 13.30 Lunch
13.30 - 15.15 Interpretable Machine Learning
15.15 - 15.30 Coffee Break
15.30 - 17.00 Special Topics in Machine Learning
approx. 18.30 Dinner

Tuesday, 9 June 2026 - Generative AI
09.00 - 10.15 Working with Generative AI: Basics and Best Practices
10.15 - 10.30 Coffee Break
10.30 - 12.30 Advanced Concepts in Generative AI
12.30 - 13.30 Lunch
13.30 - 15.00 Introduction to Agentic AI: Concepts and Applications
15.00 Concluding Remarks and Closing of Seminar (EAA)

Vorgehensweise und Ziele

The purpose of this seminar is to provide participants with a solid understanding of both traditional and modern AI technologies, blending conceptual knowledge with hands-on coding exercises. Participants will learn to critically assess, implement, and experiment with machine learning models and generative AI techniques, and apply them effectively in actuarial practice, tailored to their professional needs and interests.

By the end of the seminar, participants will be able to:

• Distinguish between traditional and modern machine learning models, evaluate their strengths and weaknesses, and apply them to actuarial problems.
• Understand the role and importance of model interpretability and ethics, and apply explainable AI techniques to enhance transparency.
• Explore special machine learning topics such as clustering, time series forecasting, and synthetic data generation, with the option to focus on topics aligned with participants’ interests.
• Apply basic and advanced prompting techniques to generative AI models, recognizing potential pitfalls and using appropriate error measures to evaluate performance.
• Integrate LLMs into actuarial workflows via APIs or local setups, leveraging advanced techniques such as function calling, fine-tuning, structured outputs, and retrieval-augmented generation.
• Understand the emerging field of agentic AI, including its concepts, practical implementations, and potential applications in insurance and finance.

Other topics that are not explicitly mentioned – such as regulatory issues, deployment of AI solutions, or governance frameworks – are not the primary focus but may be addressed where relevant during discussions or hands-on exercises.

The seminar strikes a balance between conceptual understanding, practical coding experience, and forward-looking insights, leaving participants with a toolkit of methods and techniques to responsibly and effectively apply machine learning and generative AI in their professional practice.

Teilnehmer

This seminar is designed for actuaries, data scientists, statisticians, and other professionals in the insurance and financial sectors who wish to deepen their knowledge of modern machine learning methods and generative AI. Basic programming knowledge (e.g., Python) is recommended, while familiarity with machine learning and generative AI concepts is a plus but not required. Most coding will be provided and guided by the instructor, allowing participants to experiment with, modify, and extend the provided examples to gain hands-on experience.

To fully benefit from the seminar, participants are strongly encouraged to bring a laptop. All coding will be done in Python using Jupyter notebooks, with live coding sessions forming an integral part of the programme. Exercises can be completed either via provided online platforms or on participants’ own setups, and detailed installation instructions will be shared in advance to ensure a smooth start.

Dozierende

Dr Simon Hatzesberger
Simon Hatzesberger is an actuary working as a Manager in Actuarial & Insurance Services at Deloitte. During his previous tenure in the actuarial department at Allianz Private Health, he was responsible for various data- and AI-related topics for several years. He holds an MSc degree in Financial Mathematics and Actuarial Sciences from the Technical University of Munich, as well as an MSc degree in Computer Science and a PhD in Mathematical Stochastics from the University of Passau. Additionally, he is a member of the German Association of Actuaries, a Certified Actuarial Data Scientist, and a Certified Enterprise Risk Actuary. He is actively involved in several Actuarial Data Science committees of the German Association of Actuaries, serves as a workstream lead in the Artificial Intelligence Task Force of the International Actuarial Association, and is a member of EIOPA’s Consultative Expert Group on Data Use in Insurance.

Sprache/Kurztitel

The language of the seminar will be English.

CPD Credits
For this seminar, the following CPD credits are available under the CPD scheme of the relevant national actuarial association:

• Austria: 11.5 points
• Belgium:  11 points
• Bulgaria:  15 points
• Croatia: individual accreditation
• Czechia:  11 hours
• Denmark: 12 credits
• Estonia:  11.25 hours
• Finland: 7.75 points
• France:  66 points
• Germany:  12 hours
• Greece: 15 points
• Hungary:  12 hours
• Iceland: 11.5 credits
• Ireland: 11.25 hours
• Italy: approx. 4 credits (individual accreditation)
• Latvia: 11 hours
• Lithuania: 11.5 hours
• Netherlands:  approx. 11.25 points (individual accreditation)
• Norway: 11 points
• Poland: 11.25 hours
• Portugal: 11.25 hours
• Serbia: 5 hours
• Slovakia:  8 points
• Slovenia:  50 points
• Spain: CAC: 11 hours, IAE: 11 hours
• Switzerland:  15 points
• USA: SOA (Section B): up to 13.20 hours

No responsibility is taken for the accuracy of this information.

Veranstaltungsort und Hotel

The seminar will take place at the hotel

Novotel Munich City
Hochstrasse 11
81669 Munich, Germany
Hotel website

We have arranged special prices for accommodation. The special rate for a single room is €129.00 per night, including breakfast and VAT. It is valid for bookings by 10 May 2026 out of our allotment “EAA Seminar”. Our allotment includes a limited number of rooms. Kindly book your accommodation directly with the hotel (booking form), and note the hotel’s cancellation policy.