|
Activity Number:
|
396
- Distributional Robustness, Validity, Causality, and Generalizability
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 12, 2021 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
IMS
|
|
Abstract #316891
|
|
|
Title:
|
Causality and Distribution Generalization
|
|
Author(s):
|
Jonas Peters* and Rune Christiansen and Niklas Pfister and Martin Jakobsen and Nicola Gnecco
|
|
Companies:
|
University of Copenhagen and University of Copenhagen and University of Copenhagen and University of Copenhagen and University of Geneva
|
|
Keywords:
|
causality;
distribution generalization;
interventions;
invariance;
distributional robustness
|
|
Abstract:
|
Purely predictive methods do not perform well when the test distribution changes too much from the training distribution. Causal models are known to be stable with respect to distributional shifts such as arbitrarily strong interventions on the covariates, but may not perform well when the test distribution differs only mildly from the training distribution. As a result, methods have been proposed that provide a trade-off between causal and predictive models. In this talk, we show some practical and theoretical implications of this idea and discuss theoretical limitations for the case of nonlinear models. This is joint work with Rune Christiansen, Nicola Gnecco, Martin Jakobsen, and Niklas Pfister.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2021 program
|
