Explore advanced techniques in multilingual syntactic analysis through this comprehensive lecture by Slav Petrov from Google. Delve into a coarse-to-fine architecture for dependency parsing that employs linear-time vine pruning and structured prediction cascades, resulting in a faster and more accurate model compared to traditional approaches. Learn about an online algorithm for training structured prediction models with extrinsic loss functions, demonstrating how parsing accuracy impacts downstream applications like machine translation. Discover methods for projecting part-of-speech taggers and syntactic parsers across language boundaries, enabling model building for languages lacking labeled training data. Gain insights into the speaker's work on natural language processing, machine learning, and their applications in syntactic parsing, machine translation, and information extraction.
Fast, Accurate and Robust Multilingual Syntactic Analysis - 2012
Center for Language & Speech Processing(CLSP), JHU via YouTube
Overview
Syllabus
Fast, Accurate and Robust Multilingual Syntactic Analysis – Slav Petrov (Google) - 2012
Taught by
Center for Language & Speech Processing(CLSP), JHU