Are you interested in developing cutting-edge Bayesian AI agents? In this postdoc you will design Bayesian agents that predict and advise on flexible energy usage. By reasoning with uncertainty, these agents contribute to the reliability of energy networks in the Netherlands. This research requires a multidisciplinary approach, combining probabilistic (Bayesian) AI techniques with domain knowledge from industrial partners. Please see this
video (https://youtu.be/QYbcm6G_wsk) on Natural Artificial Intelligence for more information about our research.
The Netherlands is in the middle of an energy transition in which the energy network plays a crucial role. The limited capacity of the energy network leads to net congestions that threaten the reliability and expansion of the energy network, with adverse effects on the energy transition and economic growth.
Your main task will be to design artificial agents for flexible energy management, on an individual and collective level, based on a leading physics/neuroscientific theory about computation in the brain, the Free Energy Principle (FEP). You will work closely with industry partners that specialize in energy management solutions, and with Alliander (https://www.alliander.com/en/), a leading developer and maintainer of energy networks in the Netherlands.
This postdoc position is funded by
AiNed InnovationLabs (https://ained.nl/en/current-calls/call-ained-innovatielabs-2024-stichting-ained/), which stimulates the development of new AI applications by Dutch industry and public organizations. Therefore, the project has a strong application-driven character. You will work in the
BIASlab (http://biaslab.org) team in the Electrical Engineering department at TU/e. This lab focuses its research activities on transferring the FEP to practical use in engineered solutions. During this project you will closely collaborate with other BIASlab researchers, as well as with industry partners and affiliated knowledge institutes.
Key areas of interest include Bayesian machine learning, probabilistic graphical models (factor graphs) and probabilistic programming.