Applied Mathematics Colloquia by Liam Solus: Perspectives on Causal Discovery




RE 104


Liam Solus, assistant professor of combinatorics and probability in artificial intelligence, KTH Royal Institute of Technology


Perspectives on Causal Discovery


The field of causality has recently come forth as a subject of interest in machine learning, largely due to advances in data collection methods in the biological sciences and tech industry where large-scale experimental data sets can now be efficiently and ethically obtained. The modern approach to causality decomposes the inference process into two fundamental problems: the inference of causal relations between variables in a complex system and the estimation of the causal effect of one variable on another given that such a relation exists. The subject of this talk will be the former of the two problems, which is commonly called the problem of causal discovery. In this talk, we will examine recent advances in causal discovery, starting with what can be done when we have access to a mixture of observational and experimental data. From there, we will examine emerging methods that allow us to infer casual relations even when access to experimental data is denied. Geometry, algebra and combinatorics will naturally make an appearance as we examine these methods. Time permitting, we will see how our efforts to address the problem of causal discovery lead us to answering some open questions in these more theoretical fields.


Applied Mathematics Colloquium


Event Contact

Getting to Campus