Research for Trustworthy Predictive Models

Imagine you’re measuring the temperature on your patio every day and writing it down. These are what we call “measurement data.” Some days are warmer, some are colder, but overall, it gets warm in the summer and cold in the winter. This overall pattern is called a “trend.” In our work, we focus on finding or learning these trends in the measurement data. We do this using something called Gaussian Process Models. A Gaussian Process Model is great at “learning” these trends. Even if there are a few unusually warm days in winter, the model still understands that it’s winter. Through learning, Gaussian Processes can also make predictions for the future, such as what the temperatures might be in the coming years.

Our research aims to improve these predictions and use them to detect changes and anomalies in the data. For example, we’ve been able to make Gaussian Processes learn new patterns faster. Additionally, we’ve taught the models some basic physical laws! Right now, we’re working on figuring out when a model is “good,” so we can compare different models. After all, we don’t want our temperature model to predict several warm days every winter just because it happened once.

Finally, we also want a Gaussian Process to notice when new data looks different from what it has seen before and explain what makes it different. For instance, if we put up an umbrella on our patio and it doesn’t get as warm during the day anymore, the model should say, “Warning: it’s not getting as warm during the day as before. Something has changed.”

Overall, with our project, we aim to help understand complex processes better and explain changes in those processes. By providing clear explanations, we can ensure that we can trust the predictions made by our models.

Additional resources

TBA

Cooperation

Fernuni Hagen

Data Management and Analytics Group

Prof. Dr. Christian Beecks

PhD student: Jan David HĂŒwel

Department of Electrical Engineering and Computer Science

Prof. Dr. Markus Lange-Hegermann

PhD student: Andreas Besginow

Project Publications

  • Berns, Fabian, Jan David HĂŒwel, and Christian Beecks (2021). ‘‘LOGIC: Prob- abilistic Machine Learning for Time Series Classification’’. In: 2021 IEEE International Conference on Data Mining (ICDM), pp. 1000–1005. doi: 10.1109/ICDM51629.2021.00113.
  • Berns, Fabian, Jan David HĂŒwel, and Christian Beecks (2022). ‘‘Automated Model Inference for Gaussian Processes: An Overview of State-of-the-Art Methods and Algorithms’’. In: SN Computer Science 3.4, pp. 1–11.
  • Besginow, Andreas, Jan David HĂŒwel, Markus Lange-Hegermann, and Christian Beecks (2020). ‘‘Exploring Methods to Apply Gaussian Processes in Industrial Anomaly Detection’’. In: Neurocomputing 403, pp. 383–399.
  • Besginow, Andreas and Markus Lange-Hegermann (2022). ‘‘Constraining Gaus- sian Processes to Systems of Linear Ordinary Differential Equations’’. In: Advances in Neural Information Processing Systems. Ed. by Alice H. Oh, Alekh Agarwal, Danielle Belgrave, and Kyunghyun Cho.
  • Gresch, Anne, Jan David HĂŒwel, Jennifer Briggs, Tim Berger, Ruben Koch, Thomas Deickert, Christian Beecks, Richard Benninger, and Martina DĂŒfer (2023). ‘‘Resolving spatiotemporal electrical signaling within the islet via CMOS microelectrode arrays (Under review)’’. In: bioRxiv. doi: 10.1101/ 2023.10.24.563843.
  • HĂŒwel, Jan David and Christian Beecks (2023). ‘‘Gaussian Process Compo- nent Mining with the Apriori Algorithm’’. In: International Conference on Database and Expert Systems Applications. Springer, pp. 423–429.
  • HĂŒwel, Jan David and Christian Beecks (2024). ‘‘Frequent Component Analysis for Large Time Series Databases with Gaussian Processes (under review)’’. In: International Conference on Extending Database Technology.
  • HĂŒwel, Jan David, Fabian Berns, and Christian Beecks (2021). ‘‘Automated Kernel Search for Gaussian Processes on Data Streams’’. In: IEEE BigData. IEEE, pp. 3584–3588.
  • HĂŒwel, Jan David, Andreas Besginow, Fabian Berns, Markus Lange-Hegermann, and Christian Beecks (2020). ‘‘On Kernel Search Based Gaussian Process Anomaly Detection’’. In: International Conference on Innovative Intelligent Industrial Production and Logistics. Springer, pp. 1–23.
  • HĂŒwel, Jan David, Andreas Besginow, Fabian Berns, Markus Lange-Hegermann, and Christian Beecks (2023). ‘‘On Kernel Search Based Gaussian Process Anomaly Detection’’. In: Innovative Intelligent Industrial Production and Logistics. Ed. by Alexander Smirnov, HervĂ© Panetto, and Kurosh Madani. Cham: Springer Nature Switzerland, pp. 1–23. isbn: 978-3-031-37228-5.
  • HĂŒwel, Jan David, Anne Gresch, Tim Berger, Martina DĂŒfer, and Christian Beecks (2022). ‘‘Analysis of Extracellular Potential Recordings by High- Density Micro-electrode Arrays of Pancreatic Islets’’. In: International Conference on Database and Expert Systems Applications. Springer, pp. 270–276.
  • HĂŒwel, Jan David, Anne Gresch, Fabian Berns, Ruben Koch, Martina DĂŒfer, and Christian Beecks (n.d.). ‘‘Tracing Patterns in Electrophysiological Time Series Data’’. In: International Conference on Data Science and Advanced Analytics (in print).
  • HĂŒwel, Jan David, Florian Haselbeck, Dominik G Grimm, and Christian Beecks (2022). ‘‘Dynamically Self-adjusting Gaussian Processes for Data Stream Modelling’’. In: German Conference on Artificial Intelligence (KĂŒnstliche Intelligenz). Springer, pp. 96–114.
  • Schlake, Georg Stefan, Jan David HĂŒwel, Fabian Berns, and Christian Beecks (2022). ‘‘Evaluating the Lottery Ticket Hypothesis to Sparsify Neural Net- works for Time Series Classification’’. In: 2022 IEEE 38th International Conference on Data Engineering Workshops (ICDEW). IEEE, pp. 70–73.