Events International Conference on Foundations and Advances of Machine Learning in Official Statistics

3rd – 5th April, 2024 in Wiesbaden, Germany

Call for Abstracts

Steady modernization and acceptance of new technologies and methodological developments allow official statistics to continuously offer timely and qualitative statistics to politics, the public and private sector, as well as the general public. Machine learning (ML) techniques offer an opportunity to bridge traditional statistical methods and innovative technology, thereby unlocking the potential to increase efficiency and precision in data collection and processing. At the same time, the adaption of machine learning in official statistics raises a series of demands to be fulfilled. Among them are sound and appropriate methodology, quality aspects, such as explainability and reproducibility, ethical and legal considerations and technical challenges.

The International Conference on Foundations and Advances of ML in Official Statistics is organized by the Federal Statistical Office of Germany (Destatis) as an in-person event in Wiesbaden, Germany from 3rd to 5th April, 2024. It offers a forum for international and interdisciplinary exchange on a large selection of topics around machine learning in official statistics. We invited contributions from researchers in the area of machine learning, statisticians from National Statistical Institutes (NSIs), Other National Authorities (ONAs) and inter- and supranational organizations, as well as others interested in the subject. Topics include, but are not limited to:

  • State-of-the-art methods and techniques
  • Quality aspects of machine learning
  • National and international standards and their harmonization
  • Applications of machine learning in official statistics
  • Real-time statistics facilitated by machine learning
  • Machine Learning in data editing and imputation
  • Practical experiences and pilot projects
  • Automation through machine learning
  • Legal and ethical aspects
  • Robust models
  • Interpretability and explainability of machine learning models
  • Reproducibility and long-term archives for machine learning models
  • Organizational structures, ModelOps and technical aspects
  • Scalability of ML solutions
  • Concepts for training and continuous education in machine learning

As an extension and supplement to the conference, it is planned to publish a book entitled "Foundations and Advances of ML in Official Statistics" (editor: Florian Dumpert).