@InProceedings{ruder-plank:2017:EMNLP2017,
  author    = {Ruder, Sebastian  and  Plank, Barbara},
  title     = {Learning to select data for transfer learning with Bayesian Optimization},
  booktitle = {Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing},
  month     = {September},
  year      = {2017},
  address   = {Copenhagen, Denmark},
  publisher = {Association for Computational Linguistics},
  pages     = {372--382},
  abstract  = {Domain similarity measures can be used to gauge adaptability and select
	suitable data for transfer learning, but existing approaches define ad hoc
	measures that are deemed suitable for respective tasks. Inspired by work on
	curriculum learning, we propose to learn data selection measures using Bayesian
	Optimization and evaluate them across models, domains and tasks.
	Our learned measures outperform existing domain similarity measures
	significantly on three tasks: sentiment analysis, part-of-speech tagging, and
	parsing.  We show the importance of complementing similarity with diversity,
	and that learned measures are--to some degree--transferable across models,
	domains, and even tasks.},
  url       = {https://www.aclweb.org/anthology/D17-1038}
}