Modeling AI/ML in a Federated Learning Framework

Artificial intelligence - and specifically machine learning - will be a big part of the future of aviation. Regulators are on board with the publications of roadmaps for the introduction of certified AI/ML algorithms to support air travel. Development of AI models depends critically on data. When the data used for these models resides within multiple jurisdictions, the problem of optimal – and legal – data sharing becomes even more critical. A relatively recent technical development that supports the use of distributed data sources to get to a common AI solution is called Federated Learning. IDCA is situated uniquely to develop guidance to make this happen. This guidance will consist of consensus standards that allow models to be built and validated without ever pooling all the data together. It will also ensure that all local data storage and usage laws are respected.

By downloading the issue paper, you give us permission to contact you for more information.