Semi-Supervised Guided Deep Learning to Automatically Add Semantics to Time Series

24 March 2022, Leuven & online
AI for Times Series seminar, by Peter Karsmakers (KU Leuven DTAI, Geel Campus)
VAIA, Flanders AI Research & KU Leuven STADIUS

A deep neural network model typically is learned solely from data in the form of input-output pairs, ignoring domain knowledge that additionally might be available. When domain knowledge is injected to serve as a learning guidance it is expected that the need for annotated data is relaxed and training is less dependent on the initialization of the model parameters. Domain knowledge can be provided in many different types and forms. In this seminar, domain knowledge is assumed to be given as inequality constraints which appear naturally in practice. Compared to other neuro-symbolic approaches the discussed method is also able to incorporate non-linear inequality constraints and does not require to first transform the constraints into some ad-hoc term that can be added to the learning (optimisation) objective. Furthermore, it will be explained that also a semi-supervised learning mode is possible that enables learning from unlabeled data by modifying the model weights in such a way that the constraints for the unlabeled data are satisfied. After the concepts are being introduced different use-cases in the context of automated interpretation of time series will be discussed.

Peter Karsmakers

Peter Karsmakers received the PhD degree from the Department of Electrical Engineering, KU Leuven, in 2010. In 2013, as a Postdoctoral Researcher, he co-founded the ADVISE Research Group at Geel together with a few other colleagues. Since 2018, he has been an Assistant Professor with the Computer Science Department, KU Leuven. His research interest includes designing machine learning algorithms that consider application-specific constraints. These can, for example, relate to the computing platform on which the algorithm will be deployed or physical consistency between model variables.


  • 24 March 2022, 14.30-15.30
  • Location: online streaming
  • Contact: Philippe Dreesen & Katrien De Cock
  • Language: English
  • Target audience: everyone interested in AI and/or time series


  • Price: free
  • Registration is not necessary, you can simply join the seminar.
  • However, if you would like to be reminded of this seminar, fill in your contact details below and you will receive a notification on the day of the seminar.

AI for Time Series Seminars

Several research groups in the Flanders AI Research Program (FLAIR) conduct world-class research on time series, both in the development of algorithms and tools, as in a wide area of application fields. In a recent poll in the Flanders AI community, ‘time series’ came up as the most wanted topic for future workshops or courses. With this seminar series, we bring together researchers that are interested in, or are conducting research related to, time series. We offer a varied program of national and international speakers.

Ready to get started

Remind me of the
AI for Time Series Seminars

Register here to receive reminders of the AI for Time Series Seminars.
We will use this information only for this purpose as stipulated in our Privacy Statement & Disclaimer. You can unregister at any time via the unregister link at the bottom of each reminder.