Title:	Machine learning for causal inference with applications in healthcare
DNr:	NAISS 2024/22-285
Project Type:	NAISS Small Compute
Principal Investigator:	Fredrik Johansson <fredrik.johansson@chalmers.se>
Affiliation:	Chalmers tekniska högskola
Duration:	2024-02-25 – 2025-03-01
Classification:	10201
Homepage:	http://www.fredjo.com
Keywords:

Abstract

Machine learning (ML) is becoming an integral tool in automating and improving healthcare. In particular, prediction tasks and decision support are important problems addressed using ML applied to electronic health records, genetic sequencing data and records from clinical trials. Common issues include domain adaptation, identification of causal effects and policy evaluation. My group performs machine learning research both on the methodological and applied side. For this, we require training of both deep learning and classical statistical models. In other words: not all of our projects require the use of GPUs but benefit more from parallell CPU computation.