Group Distributionally Robust Optimization for Solving Out-Of-Domain Generalization and Finding Causal Invariant Relationships

Bachelor Thesis (2022)
Author(s)

Z. Guan (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Contributor(s)

J.H. Krijthe – Mentor (TU Delft - Pattern Recognition and Bioinformatics)

R.K.A. Karlsson – Mentor (TU Delft - Pattern Recognition and Bioinformatics)

S.R. Bongers – Mentor (TU Delft - Pattern Recognition and Bioinformatics)

Thomas Höllt – Graduation committee member (TU Delft - Computer Graphics and Visualisation)

Faculty
Electrical Engineering, Mathematics and Computer Science
Copyright
© 2022 Zenan Guan
More Info
expand_more
Publication Year
2022
Language
English
Copyright
© 2022 Zenan Guan
Graduation Date
24-06-2022
Awarding Institution
Delft University of Technology
Project
CSE3000 Research Project
Programme
Computer Science and Engineering
Faculty
Electrical Engineering, Mathematics and Computer Science
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Out-of-Domain (OOD) generalization is a challenging problem in machine learning about learning a model from one or more domains and making the model perform well on an unseen domain. Empirical Risk Minimization (ERM), the standard machine learning method, suffers from learning spurious correlation in the training domain, therefore may perform badly when the unseen domain has different distribution from the training domain. Group Distributionally Robust Optimization (group DRO) is a method proposed to handle the OOD generalization problem. In this paper, the goals are to 1) measure if group DRO has a better OOD generalization performance than ERM. 2) evaluate if group DRO finds causally invariant relationships between the input and output. Semi-synthetic bird images with different backgrounds are used to form our data sets to construct a binary image classification problem for experiments. Results show that group DRO improves OOD generalization performance over ERM, and group DRO can find invariant relationships. However, the ability of group DRO to find invariant relationships is limited when the spurious correlation in the training domain is strong.

Files

License info not available