AI-based Model Fusion

The following vacant PhD position will be active in the first grand challenge of this impulse program, called Data Science: Hybrid, Automated, Trusted, and Actionable and will include the following topics: - AI-based Model fusion: At a system level, multiple predictive models are often integrated to support a decision. These models may originate from multiple inputs, outputs, scales or contexts. This task focuses on the integration of these various heterogeneous types of knowledge into a general framework optimally fusing expert-driven models as well as data-driven machine (deep) learning models. The aim is to share as much as possible common properties among inputs/outputs in order to build up knowledge on common layers. In particular, this PhD will develop advanced AI based methodology to extend decision support systems with multi-view/multitask/multi-scale model fusion in the following sense: - Multi-view: integrate many inputs coming from a variety of modalities (sensors). How to combine? How to cope with variability in the modality subsets? - Multi-task: combine many outcomes referring to different diagnostic questions. This involves integration of different models with a common knowledge layer. - Multi-scale: aggregrate many sub-models for different ‘sub-diagnostics’ in higher-level models for high-level diagnostics and validating these generic methodologies in a variety of proof-of-concept applications in Health sciences and industry.