EAA Web Session: "GenAI Beyond the Basics: Advanced Concepts for Actuaries"

Einleitung/Dauer

Generative AI (GenAI) tools, such as ChatGPT, are rapidly reshaping the way actuaries approach problem-solving, analysis, and communication. Beyond their familiar web-based interfaces, these tools offer powerful functionalities through programmatic integration, which allows users to directly access and interact with the underlying Large Language Models (LLMs) via APIs (Application Programming Interfaces). Unlike standard web-based interactions, API access enables actuaries to seamlessly integrate GenAI into their existing workflows. Furthermore, it supports scalable usage, where multiple requests can be processed simultaneously or sequentially, allowing for more efficient handling of large volumes. In this two-hour web session, participants will experience interactive live demonstrations, where advanced GenAI concepts are presented and applied to real actuarial use cases.

After introducing the basics of applying GenAI through APIs, we will explore the following advanced GenAI concepts:

• Fine-Tuning: Customizing pre-trained LLMs with domain-specific data to improve accuracy and relevance in generating responses.
• Retrieval-Augmented Generation (RAG): Combining LLMs with external data sources to produce contextually enriched and up-to-date outputs.
• Function Calling: Enabling LLMs to execute predefined operations, such as calculations or database queries, by invoking specific functions in the underlying programming language.
• Structured Outputs: Generating responses in predefined formats, such as JSON or tabular data, to ensure consistency and integration with other systems or processes.

For each concept, the session will present its purpose, principles, and functionality, accompanied by a relevant actuarial use case to illustrate its practical application. The session will conclude with an outlook on future developments, followed by a space for questions and open discussion.

Vorgehensweise und Ziele

The aim of this web session is to provide participants with the knowledge and practical skills needed to apply advanced GenAI techniques in actuarial contexts. By the end of the session, participants will be able to:

• Understand how to interact with GenAI tools programmatically through APIs.
• Customize LLMs using Fine-Tuning techniques to address domain-specific requirements.
• Combine external data with LLMs through Retrieval-Augmented Generation to produce contextually relevant outputs.
• Leverage Function Calling capabilities to trigger the execution of predefined functions in the underlying programming language, such as calculations or database queries.
• Generate structured outputs in formats like JSON or tables to ensure compatibility with actuarial processes and systems.
• Apply Python programming skills to develop and integrate GenAI-powered solutions into actuarial workflows.

This session strikes a balance between theory and practical programming, equipping participants with tools to enhance the accuracy, efficiency, and creativity of their actuarial work.

Teilnehmer

This web session is designed for actuaries, data scientists, and other professionals in the insurance and financial industries looking to enhance their technical understanding of GenAI tools. It offers practical insights for beginners in programming and GenAI, while providing experienced users with the opportunity to refine their skills and master advanced techniques.

While a basic familiarity with Python (or another modern programming language) is helpful, it is not a prerequisite. The session includes clear guidance and hands-on examples to ensure accessibility for all participants.

Technical Requirements
Please check with your IT department if your firewall and computer settings support web session participation (the programme Zoom will be used for this online training). Please also make sure to join the web session with a stable internet connection.

Translation
We are excited to introduce Wordly, an AI-powered live translation tool, in this web session. Follow along with automated subtitles in multiple languages and experience how AI is shaping not just actuarial work, but also learning itself!

Dozierende

Dr Simon Hatzesberger

Simon Hatzesberger is an actuary working as a Manager in Actuarial & Insurance Services for 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. Besides his professional work, he lectures in mathematics at the Universities of Applied Sciences in Munich and Regensburg. He is actively engaged in the “Explainable Artificial Intelligence” working group of the German Association of Actuaries and is a member of the “Artificial Intelligence” task force of the International Association of Actuaries.

Sprache/Kurztitel

The language of the web session will be English.