Screenprint zoutintrusie

On Wednesday, December 13, from 11 a.m. to noon, the DigiShape AI working group is hosting an online technical session with Paula Lambregts from HKV on Explainable AI. 

More insight and confidence in predictions generated with Machine Learning

Machine learning models are built and deployed more and more frequently for all sorts of applications in the water sector. Often, they give better outcomes due to their flexibility and predictive power. However, this comes at a cost of less understanding of the system due to the black box nature of these algorithms. For this reason, users of AI (prediction) models often have limited confidence in the model output even if they perform better than their physics-based counterparts. We encounter this during our work, for example in a project with Rijkswaterstaat where we have built a Machine Learning model to predict the salinity on the Amsterdam Rhine Canal.

Our goal is to give users in the water sector more insight and confidence in predictions generated with Machine Learning models by applying Explainable AI techniques (XAI). This will be done based on an example case for the prediction of salt concentrations on the Amsterdam-Rhine Canal (ARK) in collaboration with Rijkswaterstaat. A theoretical framework for multiple XAI techniques exists, but it has not yet been put into practice within the water sector.

Practical information

  • What: Online technical session
  • When: Wednesday, December 13, from 11 to 12 a.m.
  • Location: Microsoft Teams
  • Register here

Door nieuwe technologische ontwikkelingen is het zeker dat we op een andere manier gaan varen. Maar hoe dat er precies uitziet, en welke gevolgen dat heeft voor onze infrastructuur, daar is nog geen eenduidig beeld over. Om hier een gezamenlijke aanpak voor te ontwikkelen, moeten we heel nauw samenwerken als watersector. Dat doen we binnen DigiShape.

Tom van der Schelde (RWS), programmalijn trekker Water en Logistiek

Programmalijn Water en Logistiek