"Rational Recurrences for Empirical Natural Language Processing"

Despite their often-discussed advantages, deep learning methods largely disregard theories of both learning and language.  This makes their prediction behavior hard to understand and explain.  In this talk, I will present a path toward more understandable (but still "deep") natural language processing models, without sacrificing accuracy.  Rational recurrences comprise a family of recurrent neural networks that obey a particular set of rules about how to calculate hidden states, and hence correspond to parallelized weighted finite-state pattern matching.  Many recently introduced models turn out to be members of this family, and the weighted finite-state view lets us derive some new ones.  I'll introduce rational RNNs and present some of the ways we have used them in NLP.  My collaborators on this work include Jesse Dodge, Hao Peng, Roy Schwartz, and Sam Thomson.   

https://events.cornell.edu/event/lmss_cornell_tech_noah_smith_university_of_washington?utm_campaign=widget&utm_medium=widget&utm_source=Cornell

BIO

Noah Smith is a Professor in the Paul G. Allen School of Computer Science & Engineering at the University of Washington, as well as a Senior Research Manager at the Allen Institute for Artificial Intelligence.  His research interests include statistical natural language processing, machine learning, and applications of natural language processing, especially to the social sciences. His book, Linguistic Structure Prediction, covers many of these topics. Alumni of his research group, Noah's ARK, are international leaders in NLP in academia and industry; in 2017 UW's Sounding Board team won the inaugural Amazon Alexa Prize. Smith's work has been recognized with a UW Innovation award (2016–2018), a Finmeccanica career development chair at CMU (2011–2014), an NSF CAREER award (2011–2016), a Hertz Foundationgraduate fellowship (2001–2006), numerous best paper nominations and awards, and coverage by NPR, BBC, CBC, New York Times, Washington Post, and Time.