Privacy-Preserving Machine Learning for Synthetic Spatio-Temporal Trajectory Data Generation

PI: Fredrik Heintz (LiU)

A major open research challenge is developing privacy-preserving machine learning methods that both achieves high performance and privacy guarantees even though the original training data contains sensitive personal information. The applications are abundant, from making cities safer, via on-demand public transportation systems to improved medical diagnosis.

The goal of the project is to develop new machine learning methods for creating synthetic spatio-temporal trajectory data sets preserving the privacy of the individuals in the original data. The project will 1) extend generative adversarial network (GAN) methods to learn generative spatio-temporal trajectory models and 2) develop new Bayesian Optimization methods for creating tailored privacy-preserving synthetic data sets using these generative models.

The project has access to unique trajectory data of people, busses and trains through collaborations with organizations such as Telia, Trafikverket and Östgötatrafiken. These organizations are also very interested in applying the results of the research.

The project will fund a postdoc and lies in topic A with strong connections to topics B and E in the ELLIIT 2030 Technology Forecast with important applications in themes 1 and 2. It complements and significantly extends ongoing research and will further strengthen the existing research collaborations with Lund University.

Project number: A10