We propose a mixture-of-experts approach for unsupervised domain adaptation from multiple sources. The key idea is to explicitly capture the relationship between a target example and different source domains. This relationship, expressed by a point-to-set metric, determines how to combine predictors trained on various domains. The metric is learned in an unsupervised fashion using meta-training. Experimental results on sentiment analysis and part-of-speech tagging demonstrate that our approach consistently outperforms multiple baselines and can robustly handle negative transfer.
@article{arxiv.1809.02256,
title = {Multi-Source Domain Adaptation with Mixture of Experts},
author = {Jiang Guo and Darsh J Shah and Regina Barzilay},
journal= {arXiv preprint arXiv:1809.02256},
year = {2018}
}