Upcoming seminars

Thursday, March 13, 15:00

Johannes Textor
Radboud University
Expert-In-The-Loop Causal Structure Learning

This presentation is part of the International Seminar Series in Causal Inference. The aim of the seminar series is to bring distinguished causal inference speakers to Copenhagen and to foster new connections among local causal inference researchers across different disciplines and institutions. The seminar is therefore accompanied with three additional opportunities for connections:

1. A reception following the presentation. We encourage all to participate, no registration needed.
2. A lunch with the speaker only for PhD students on March 13. If you are a PhD student interested in participating, please sign up by contacting Marie Pramming (). Note that there are limited seats.
3. A possibility to book a one-on-one meeting with the speaker on March 12 or March 13. If you are interested in this, please contact Anne Helby Petersen ().

The seminar is organized by the Pioneer Centre for SMARTBiomed and supported by the Danish Data Science Academy.

Abstract: Numerous causal discovery algorithms were developed to automatically learn directed acyclic graphs (DAGs) and other causal models from data. However, their adoption in applied domains remains limited, as researchers often prefer to construct DAGs manually based on domain knowledge. This preference arises due to several practical challenges with automated algorithms, such as their tendency to produce results that contradict obvious domain knowledge and their inability to distinguish Markov equivalent models. To assist researchers in constructing DAGs manually, we propose an iterative structure learning approach that combines domain knowledge with data-driven insights. Our method leverages conditional independence testing to iteratively identify variable pairs where an edge is either missing or superfluous. Based on this information, we can choose to add missing edges with appropriate orientation based on domain knowledge or remove unnecessary ones. We also give a method to rank these missing edges based on their impact on the overall model fit. In a simulation study, we find that this iterative approach to leverage domain knowledge already starts outperforming purely data-driven structure learning if the orientation of new edge is correctly determined in at least two out of three cases. We present a proof-of-concept implementation using a large language model as a domain expert and a graphical user interface designed to assist human experts with DAG construction.

Monday, March 24, 10:00

Morten Mørup
Danish Technical University
Characterizing Neural Responses, Individual Variability, and the Functional and Structural Organization of Connectomes using Machine Learning

This talk will describe our research efforts towards using machine learning to extract patterns and account for individual variability in functional neuroimaging data. A primer will be given to machine learning and the importance of model generalization as well as how (non-parametric) Bayesian inference and predictive assessments can be used to evidence plausible hypothesis of brain organization. The talk will further highlight efforts towards characterizing individual variability in neural responses by use of deep learning voice conversion technologies adapted to the functional neuroimaging domain. Finally it will be discussed how machine learning can potentially be used to reduce the number of trials necessary in functional neuroimaging research.

Tuesday, March 25, 10:00

Benoit Liquet-Weiland
Macquarie University, School of Mathematical and Physical Sciences, Australia
Sparse Group Variable Selection for Pleiotropic Association in High-Dimensional Genomic Context

Genome-wide association studies (GWAS) have identified genetic variants associated with multiple complex diseases. We can leverage this phenomenon, known as pleiotropy, to integrate multiple data sources in a joint analysis. Often, integrating additional information such as gene pathway knowledge can improve statistical efficiency and biological interpretation. In this talk, I will review several frequentist and Bayesian statistical methods we have developed, which incorporate both gene pathway and pleiotropy knowledge to increase statistical power and identify important risk variants affecting multiple traits. Our methods are applied to identify potential pleiotropy in an application considering the joint analysis of thyroid and breast cancers.

Map of CSS

You can find CSS next to the Botanical Garden, 5 minutes from Nørreport station.


Meeting room 5.2.46 is the library of the Biostatistics section, located in building 5, 2nd floor, room 46. See the map below for directions inside CSS.