Federica Kulka

PhD
Vrije Universiteit Amsterdam (VU)

Our position is on neurosymbolic world models - predictive models that combine latent and explicit representations and reasoning methods to simulate world states in a multimodal setting. Such models will provide an interactive environment that agents can use for data-efficient exploration, planning, counterfactual reasoning, and learning. This project connects our expertise in neurosymbolic AI, commonsense reasoning, foundation models, and probabilistic ML. This position is embedded within the G-Research funding program for ELLIS Amsterdam.

Industry Track
October 1st, 2026 - September 30th, 2030
ELLIS Edge Newsletter
Join the 6,000+ people who get the monthly newsletter filled with the latest news, jobs, events and insights from the ELLIS Network.