Walter Nelson

PhD
Institute of Science and Technology Austria (ISTA)
Learning and exploiting causal relationships: from theory to application

Machine learning models are increasingly deployed in complex, high-stakes environments. Imbuing models with the ability to represent the causal structure of such complex environments has the potential to improve the quality of predictions, enhance the transparency of otherwise black-box models, and enable models to be deployed where it might otherwise be impossible. Fortunately, with complexity comes information, and we postulate that this information can be exploited by our models to learn useful causal relations with minimal human supervision. My interests lie in algorithms for learning and using such causal relations, with a particular focus on models that are practical for application in domains such as biomedicine.

Track:
Academic Track
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.