01 Dec 2014

In the lab meeting this week, on Wed Dec 3ed, we will be hosting Stefan Riezler from Heidelberg University. He is going to give a talk about Response-Based Online Structured Prediction (with Applications to Grounded Statistical Machine Translation). The meeting will be at TASC1 9204 West 10:30am.

ABSTRACT: In response-based structured prediction, instead of a gold-standard structure, the learner is given a response to a predicted structure. Different types of environments such as an extrinsic task, a computer program, or a human, can respond in form of a ranking, an acceptance/rejection decision, or an improvement of the predicted structures. In this talk, we present three instantiations of response-based learning scenarios for grounded statistical machine translation (SMT), where response signals are elicited by embedding SMT into cross-lingual information retrieval, into multilingual database access, and into human corrections of translations.

BIO: Prof. Stefan Riezler has been appointed full professor and head of the chair of Linguistic Informatics at Heidelberg University in 2010, after spending a decade in the world’s most renowned industry research labs (Xerox PARC, Google). He received his PhD in Computational Linguistics from the University of Tübingen in 1998, and then conducted post-doctoral work at Brown University in 1999. Prof. Riezler’s research focus is on machine learning and statistics applied to natural language processing problems, especially for the application areas of natural-language based web search and statistical machine translation.