Pseudo-Labeling for Unsupervised Domain Adaptation with Kernel GLMs

Nathan Weill Kaizheng Wang · 2026

Read Pseudo-Labeling for Unsupervised Domain Adaptation with Kernel GLMs by Nathan Weill Kaizheng Wang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-19422.txt.

Introduction

Weill and Wang adapt kernel models to a new domain when only source data carry labels. Their plan splits the labeled source set into two parts for two jobs. One part trains a family of candidate models. The other builds an imputation model that makes pseudo-labels for target data. Those labels then support sound model choice for the target. Risk bounds state the cost of adaptation via an effective labeled size. Tests on fake and real data beat source-only baselines with steady gains.

Catalog record

CREATORNathan Weill Kaizheng Wang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE164579
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.19422

Browse the full library · About the Tower of Babel