Dissertation/Thesis Abstract

Optimizing Machine Translation by Learning to Search
by Galron, Daniel, Ph.D., New York University, 2012, 218; 3546405
Abstract (Summary)

We present a novel approach to training discriminative tree-structured machine trans- lation systems by learning to search. We describe three primary innovations in this work: a new parsing coordinator architecture and algorithms to generate the required training examples for the learning algorithm; a new semiring that provides an unbiased way to compare translations; and a new training objective that measures whether a translation inference improves the quality of a translation. We also apply the reinforcement learning concept of exploration to SMT. Finally, we empirically evaluate our innovations.

Indexing (document details)
Advisor: Melamed, Dan
Commitee: Bangalore, Srinivas, Grishman, Ralph, Mohri, Mehryar, Sekine, Satoshi
School: New York University
Department: Computer Science
School Location: United States -- New York
Source: DAI-B 74/04(E), Dissertation Abstracts International
Source Type: DISSERTATION
Subjects: Computer science
Keywords: Computational linguistics, Machine learning, Machine translation, Natural language processing
Publication Number: 3546405
ISBN: 9781267799647
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest