A New PAC-Bayesian Perspective on Domain Adaptation

Abstract

We study the issue of PAC-Bayesian domain adaptation: We want to learn, from a source domain, a majority vote model dedicated to a target one. Our theoretical contribution brings a new perspective by deriving an upper-bound on the target risk where the distributions’ divergence - expressed as a ratio - controls the trade-off between a source error measure and the target voters’ disagreement. Our bound suggests that one has to focus on regions where the source data is informative. From this result, we derive a PAC-Bayesian generalization bound, and specialize it to linear classifiers. Then, we infer a learning algorithm and perform experiments on real data.

Cite

Text

Germain et al. "A New PAC-Bayesian Perspective on Domain Adaptation." International Conference on Machine Learning, 2016.

Markdown

[Germain et al. "A New PAC-Bayesian Perspective on Domain Adaptation." International Conference on Machine Learning, 2016.](https://mlanthology.org/icml/2016/germain2016icml-new/)

BibTeX

@inproceedings{germain2016icml-new,
  title     = {{A New PAC-Bayesian Perspective on Domain Adaptation}},
  author    = {Germain, Pascal and Habrard, Amaury and Laviolette, François and Morvant, Emilie},
  booktitle = {International Conference on Machine Learning},
  year      = {2016},
  pages     = {859-868},
  volume    = {48},
  url       = {https://mlanthology.org/icml/2016/germain2016icml-new/}
}