Invited speaker

Alexander Fraser
CIS, LMU Munich

Translation to Morphologically Rich Languages

While statistical techniques for machine translation have made significant progress in the last 20 years, results for translating to morphologically rich languages are still mixed. In particular, current research on translating to morphologically rich languages varies greatly in the amount of linguistic knowledge used and the form of this linguistic knowledge. The talk will discuss state-of-the-art techniques for translation tasks involving translating to a target language which is morphologically richer than the source language.

Speaker bio

Alexander Fraser leads the machine translation group at the Center for Information and Language Processing (CIS), LMU Munich. Previously, he led the statistical machine translation group at the Institute for Natural Language Processing (IMS) at the University of Stuttgart. He is the holder of a European Research Council Starting Grant, PI of a German Research Foundation project which is on modeling morphosyntactic phenomena in translation, and a PI in the Health in my Language (HimL) Horizon2020 Innovation Action. His main research interests are in data-driven and hybrid approaches to machine translation, domain adaptation for structured prediction, syntactic parsing, and information retrieval