Institut für Computerlinguistik,
Im Neuenheimer Feld 325
69120 Heidelberg, Germany
|Office hours||Wednesdays, 13:00–14:00
|Phone||+49 6221 54-3245 (Secretary)|
Amazon and University of Heidelberg organize a Shared Task on Bandit Learning for Machine Translation with the goal to encourage researchers to investigate algorithms for learning from weak user feedback instead of from human references or post-edits that require skilled translators. We are interested in finding systems that learn efficiently and effectively from this type of feedback, i.e. they learn fast and achieve high translation quality. Developing such algorithms is interesting for interactive machine learning and for human feedback in NLP in general. See the WMT17 page for more information.
A new research project on auto-adaptive learning from weak feedback for interactive machine lecture translation will soon start. The project attempts to learn machine translation from bandit feedback in form of judgements on the quality of a predicted translation without requiring a post-edit or a gold-standard translation. The application scenario is the translation of university lectures, a focus which is shared with parallel projects at RWTH Aachen, KIT Karlsruhe, and the University of the Saarland, Saarbrücken.