A Theory of Label Propagation for Subpopulation Shift

02/22/2021
by   Tianle Cai, et al.
4

One of the central problems in machine learning is domain adaptation. Unlike past theoretical work, we consider a new model for subpopulation shift in the input or representation space. In this work, we propose a provably effective framework for domain adaptation based on label propagation. In our analysis, we use a simple but realistic “expansion” assumption, proposed in <cit.>. Using a teacher classifier trained on the source domain, our algorithm not only propagates to the target domain but also improves upon the teacher. By leveraging existing generalization bounds, we also obtain end-to-end finite-sample guarantees on the entire algorithm. In addition, we extend our theoretical framework to a more general setting of source-to-target transfer based on a third unlabeled dataset, which can be easily applied in various learning scenarios.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/10/2020

Domain Adaptation with Conditional Distribution Matching and Generalized Label Shift

Adversarial learning has demonstrated good performance in the unsupervis...
research
02/03/2023

Domain Adaptation via Rebalanced Sub-domain Alignment

Unsupervised domain adaptation (UDA) is a technique used to transfer kno...
research
11/29/2020

Importance Weight Estimation and Generalization in Domain Adaptation under Label Shift

We study generalization under label shift in domain adaptation where the...
research
08/04/2020

Learning from a Complementary-label Source Domain: Theory and Algorithms

In unsupervised domain adaptation (UDA), a classifier for the target dom...
research
06/28/2023

Efficient and Multiply Robust Risk Estimation under General Forms of Dataset Shift

Statistical machine learning methods often face the challenge of limited...
research
02/26/2020

Understanding Self-Training for Gradual Domain Adaptation

Machine learning systems must adapt to data distributions that evolve ov...
research
10/06/2022

Improving the Sample Efficiency of Prompt Tuning with Domain Adaptation

Prompt tuning, or the conditioning of a frozen pretrained language model...

Please sign up or login with your details

Forgot password? Click here to reset