Neural networks are very powerful to draw conclusions out of a massive amount of data. Especially when it comes to structuring unstructured data, neural networks already play an important role in the insurance industry, now and especially in the future.

This seminar continues the series of EAA events on modern statistical methods and their application in the insurance industry.

Organised by the EAA – European Actuarial Academy GmbH.

Participants should have basic previous knowledge in stochastics and probability theory. No specific statistical knowledge is required.

During this seminar, you will need a laptop with Google Chrome or Mozilla Firefox installed.

Oliver Mitevski
Oliver Mitevski is working as a data scientist at Munich Re. He consults Munich Re’s business units by developing analytical solutions across the whole Reinsurance Group. Previously he worked at Xaxis GmbH as a data scientist in online marketing. He holds a Master of Science in Computer Science from the University of Freiburg, Germany.

Janek Thomas
Janek Thomas studied statistics with a minor in computer science at the Ludwig-Maximilians-University in Munich. He is currently a second year PhD in the working group computational statistics at the same university. The focus of his research is on hyperparameter configuration of machine learning models using bayesian optimization. Furthermore he is interested in gradient boosting, benchmarking of machine learning methods, high-performance computing and deep neural networks. Since 2015 he is a freelance trainer for R programming with the Münchner R Kurse and co-founder of the Munich Datageeks e.V., a non-profit organization to promote data science in Munich.