Marine Data Science

Research group at the University of Rostock, Institute for Visual and Analytic Computing
We are working on machine learning for multi-modal and heterogeneous data

  • Machine Learning for Time Series and Geospatial Data

    Developing Deep Learning methods for time series data (e.g., wearable sensors) or geospatial data. Specifically interested in reliable and probabilistic models that know what they don’t know.

  • Machine Learning for Multi-modal and Tabular Data

    Developing Deep Learning methods that can learn from and make predictions with mixed tabular and sensor data, including high-cardinality categorical data

  • Real-World Data Science

    Applying the methods developed by our group to challenging data science tasks in marine ecology, underwater technology, or sensor-based activity recognition