Port-Hamiltonian (pH) systems are used in practical applications like network modeling and soft robot control. They have desirable physical properties but lack closed-form solutions, resulting in high-dimensional models requiring extensive computational resources. To accelerate computations for pH systems, we use model reduction (MOR). However, classical MOR methods may not provide sufficient approximation quality, so we investigate an approach called MOR on manifolds. Furthermore, it is crucial that the reduced-order model (ROM) retains the same physical characteristics as the high-dimensional pH system. Therefore, there is a need to develop structure-preserving MOR on manifolds for pH systems to provide computationally efficient ROMs.

Digital platforms efficiently mediate demand and supply of labour on a large scale creating digital marketplaces for services. Simultaneously, evidence shows that platforms contribute to a degradation of labour. Platforms do not guarantee minimum wage nor income security and challenge occupational health standards which are particularly important in the context of a worldwide pandemic. Nonetheless, platforms have offered marginalized workers such as migrants, racialized minorities and workers with familial obligations (often women), alternative channels for subsistence. This raises questions not only about how platforms (re)produce existing labour market inequalities related to migration status, ethnicity/race and gender, but also how labour and data mobilities intertwine with state and international labour regulations, prohibition of labour exploitation and migration control. Utilizing qualitative research methods (e.g. interviews and participant observations), NeMo investigates the impact of digitalised platforms on labour relations by zooming in on the particular sector of domestic work. A sector especially vulnerable to labour exploitation due to its legal disenfranchisement and siting in the private household. The project will reveal how inequalities related to migration status, race/ethnicity and gender are reproduced, reconfigured and resisted in platform labour. Accordingly, policy recommendations on best practices protecting workers in domestic labour platforms will be formulated.

There is a strong positive association between tobacco and cannabis use on the one hand, and mental health disorders attention-deficit/hyperactivity disorder (ADHD) and bipolar disorder on the other. This association may arise through shared risk factors (genetic and/or environmental). Alternatively, there may be causal effects such that mental health disorders lead to higher tobacco/cannabis use, or tobacco/cannabis use negatively impact mental health. Some epidemiological evidence for both causal mechanisms exists, but novel methods are needed to firmly establish direction of causality. Mendelian randomization (MR) is an exciting, innovative technique that was proposed in 1986 but only recently became viable following advances in the genetics field. MR utilizes genetic variant(s) strongly associated with a predictor variable, as proxy, or ‘instrument’, to test causal effects on an outcome variable. Confounding and reverse causation are thereby minimized. In this Rubicon project I will apply MR in large samples from international cohort studies. The host institute where the project will take place is renowned for its pioneering role in the field of causal inference and particularly MR. The results will have profound implications for our understanding of the aetiology of mental health disorders, and development of preventive interventions and clear, evidence-based public health messages.