Seminar in applied mathematics and statistics
SPEAKER: Thijs van Ommen (Universiteit van Amsterdam).
TITLE: Domain Adaptation Using Causal Inference
ABSTRACT: An important goal common to domain adaptation and causal inference is to make accurate predictions when the distributions for the source (or training) domain(s) and target (or test) domain(s) differ. In many cases, these different distributions can be modeled as different contexts of a single underlying system, in which each distribution corresponds to a different perturbation of the system, or in causal terms, an intervention. We focus on a class of such causal domain adaptation problems, where data for one or more source domains are given, and the task is to predict the distribution of a certain target variable from measurements of other variables in one or more target domains. A naive approach to these problems may incur arbitrarily large risk. We propose an approach that exploits causal inference to find invariant conditional distributions, without relying on prior knowledge of the causal graph, the type of interventions or the intervention targets. We demonstrate our approach by evaluating a possible implementation on simulated and real world data.
Tea and chocolate will be served in room 04.4.19 after the seminar.
----------
Upcoming events:
Wednesday, November 14 at 11.15: Pariya Behrouzi
Friday, November 23 at 13.15: Benjamin Christoffersen
Wednesday, November 28 at 15.15: Shota Katayama
Friday, December 14 at 14.15: Moritz M. Schauer
Friday, February 8, 2019, at 14.15: Massimiliano Tamborrino
Friday, February 15, 2019 at 14.15: Irene Tubikanec