New paper: Generative linguistics and neural networks at 60

From Joe Pater

Update: Revised version posted Dec. 30, 2017.

I’ve just finished a paper, “Generative linguistics and neural networks at 60: foundation, friction, and fusion”. I’d very much welcome your comments, either publicly here as blog comments, or privately by other means. (Note that comments may require admin approval, and if so will not show up immediately). We’ll be discussing this paper in our next Computational Linguistics Community (CLiC) meeting Oct. 27th at 10 am in ILC N451.

http://people.umass.edu/pater/pater-perceptrons-and-syntactic-structures-at-60.pdf

Abstract. The birthdate of both generative linguistics and neural networks can be taken as 1957, the year of the publication of seminal work by both Noam Chomsky and Frank Rosenblatt. This paper traces the development of these two approaches to cognitive science, from their largely autonomous early development in their first thirty years, through their collision in the 1980s around the past tense debate (Rumelhart and McClelland 1986, Pinker and Prince 1988), and their integration in much subsequent work up to the present, 2017. Although these traditions are often presented as in opposition to one another, such a presentation assumes polar versions of each approach, and ignores the ever-growing body of results that have been achieved through integration.

3 thoughts on “New paper: Generative linguistics and neural networks at 60

Leave a Reply

Your email address will not be published. Required fields are marked *