Freitag, 26.11.21, 12:00-13:00, online: Zoom
Knightian uncertainty in Financial Markets
Freitag, 3.12.21, 12:00-13:00, online: Zoom
In this talk we revisit uncertainty in probability when the underlying probability measure can not be estimated in a reliable way which is often the case in financial markets. We will see some applications where upper and lower bounds are of interest which lead to non-linear expectation operators in contrast to the very familiar and well-known linear expectation.
Sharp adaptive similarity testing with pathwise stability for ergodic diffusions
Freitag, 17.12.21, 12:00-13:00, online: Zoom
Suppose we observe an ergodic diffusion with unknown drift. We develop a fully data-driven nonparametric test for the null hypothesis that the drift is similar to a reference drift under supremum loss. Our procedure turns out to be asymptotically optimal in both rate and constant. Moreover, we investigate its behavior if the true process was driven by a fractional Brownian motion with Hurst index close to 1/2.
Therapeutic genome editing and its need for artificial intelligence
Freitag, 14.1.22, 12:00-13:00, online: Zoom
Deep generative approaches for omics data: interpretability and sample-size constraints
Freitag, 21.1.22, 12:00-13:00, online: Zoom
Deep generative models (DGMs) are promising tools, e.g., for learning latent structure in high dimensional omics data such as single-cell RNA-Seq data as well as for generating synthetic observations, e.g. for securely sharing single nucleotide polymorphism (SNP) data. Here I address the interpretability of DGMs, specifically by showing how to\nlink latent space information with observed variables (e.g. expression levels of genes). In addition I address the performance of DGMs under sample size constraints which are frequently observable when working with omics data in the biomedical context.
Freitag, 28.1.22, 12:00-13:00, online: Zoom
How much of a covariance is causal?
Freitag, 11.2.22, 12:00-13:00, online: Zoom
Can we quantify how much of the covariance between two variables is due to the causal effects of one variable on the other? I will introduce new approaches to this problem, drawing on recent advances in the theory of causal inference. As an application, I consider the relationships between an individual’s traits and their fitness in the context of evolutionary biology. By analysing such relationships casually, we can explain why certain traits evolve over time.