Label Ranking Under Ambiguous Supervision for Learning Semantic Correspondences

Abstract

This paper studies the problem of learning from ambiguous supervision, focusing on the task of learning semantic correspondences. A learning problem is said to be ambiguously supervised when, for a given training input, a set of output candidates is provided with no prior of which one is correct. We propose to tackle this problem by solving a related unambiguous task with a label ranking approach and show how and why this performs well on the original task, via the method of task-transfer. We apply it to learning to match natural language sentences to a structured representation of their meaning and empirically demonstrate that this competes with the state-of-the-art on two benchmarks.

Cite

Text

Bordes et al. "Label Ranking Under Ambiguous Supervision for Learning Semantic Correspondences." International Conference on Machine Learning, 2010.

Markdown

[Bordes et al. "Label Ranking Under Ambiguous Supervision for Learning Semantic Correspondences." International Conference on Machine Learning, 2010.](https://mlanthology.org/icml/2010/bordes2010icml-label/)

BibTeX

@inproceedings{bordes2010icml-label,
  title     = {{Label Ranking Under Ambiguous Supervision for Learning Semantic Correspondences}},
  author    = {Bordes, Antoine and Usunier, Nicolas and Weston, Jason},
  booktitle = {International Conference on Machine Learning},
  year      = {2010},
  pages     = {103-110},
  url       = {https://mlanthology.org/icml/2010/bordes2010icml-label/}
}