Category Archives: Computational linguistics

First meeting of the Computational Humanities Initiative with Laure Thompson

The first meeting of the Computational Humanities Initiative was held Friday November 18th in N400 of the Linguistics Department. Laure Thompson of the Manning College of Information and Computer Sciences presented “Computational Humanities and Human-Centered Machine Learning” to an audience that included UMass faculty and students from the departments of Classics, Philosophy, Economics, and Linguistics as well as from CICS, Nursing and SPHSS, and from Amherst and Mount Holyoke Colleges. The event was sponsored by the College of Humanities and Fine Arts and the Computational Social Science Institute. The Computational Humanities Initiative is being organized by faculty in CHFA and CICS interested in building new connections between our colleges in research and teaching. To find out about future events, join the Computational Humanities mailing list by emailing Joe Pater (pater@umass.edu).

A recording of the talk is available to those with a UMass account, and an abstract and bio are below.

Brendan O’Connor of CICS/CSSI, and Laure Thompson

The meeting participants introducing themselves.

Thompson talk abstract: Machine learning (ML) is typically used to replicate some human activity. Given a set of inputs, a system is built to produce the same outputs as a human would; thus reducing human interaction through automation. In contrast to this standard use, the computational humanities typically use ML as a tool to enable human interaction in the form of human interpretation. This alternative use centers iterative, expert human use to study humanities collections in order to gain meaningful insights from and recognize the true complexities of cultural phenomena. In this talk, I will argue that these two uses are fundamentally different paradigms of user intention. I will illustrate the characteristics of these two paradigms using two case studies drawn from the computational humanities: a Dadaist “reading” of Dada and a largescale study of the themes in science fiction.

Bio: Laure Thompson is an assistant professor in the Manning College of Information & Computer Sciences at UMass Amherst, whose research bridges machine learning and natural language processing with humanistic scholarship. Centered on humanities applications, her research focuses on understanding what computational models actually learn and how we can intentionally change what they learn. Given this humanities focus, she works with a wide range of cultural heritage corpora: from texts of science fiction novels and medieval manuscripts to images of avant-garde journals and magical gems from the ancient Mediterranean. She completed her PhD in Computer Science at Cornell University in 2020.

New online computational linguistics offering

The Department of Linguistics at the University of Massachusetts Amherst is offering a new on-line computational linguistics course, Ling 492B, Computational linguistics: Use and meaning. The course is asynchronous, so students can watch the lectures and complete the exercises at a time of the day and week that is convenient for them. It is taking place in the UMass spring semester, beginning Jan. 25th and ending May 12th. A course description and list of prerequisites follows.

To register, go to: https://www.umass.edu/uww/class-enrollment. The cost of the course, including registration fee, is $1496. It is a regular 3-credit advanced undergraduate course, and may be eligible for transfer credit, subject to your school’s approval. 

If you have any questions about the course, contact Brandon Prickett, the instructor, at brandonlprickett@gmail.com. If you have any questions about course registration, contact  the UMass Amherst UWW registration office at regoff@uww.umass.edu.

Computational linguistics: Use and meaning 

This course is an introduction to computational linguistics, the study of natural language from a computational perspective. Computational linguistics encompasses both applied (engineering) and theoretical (cognitive) issues, and in this course you will get a taste of both. You will learn how to write code to implement key algorithms for processing and analyzing linguistic structure in language corpora. You will learn how formal language models (grammars) can be implemented computationally and used to represent linguistic structure at various levels. You will use these formal language models to automatically analyze linguistic data, and see how these models can be trained using language corpora. A major focus of the course will be on statistical techniques, especially Bayesian inference, because this forms the foundation of much current work in computational linguistics, both theoretical and applied.

The course prerequisites are some familiarity with Python (functions, lists, dictionaries, file I/O, and command line) and some background in linguistics (at least one course in e.g. syntax, phonology, or general linguistics). In addition, a course in probability or statistics is recommended.

Franklin Institute Symposium in Honor of Barbara Partee (April 19th)

We are extremely happy to announce that, in honor of Professor Barbara Partee receiving the 2021 Benjamin Franklin Medal in Computer and Cognitive Science, the Franklin Institute and the University of Pennsylvania are organizing a special symposium honoring her and her legacy in the field.

Due to the COVID-19 pandemic, this symposium will be held remotely, and can be viewed publicly over Zoom. It will take place on Monday, April 19th, from 9:45AM to 3PM (EST), and will feature presentations by:

  • Barbara Partee (UMass Amherst)
  • Gennaro Chierchia (Harvard University)
  • Pauline Jacobson (Brown University)
  • Florian Schwarz (University of Pennsylvania)
  • Seth Cable (UMass Amherst)
  • Christopher Potts (Stanford University)

The website for the symposium, which includes the full program (with abstracts) as well as the Zoom link for the remote presentations, can be found at the link below:

Again, this event is entirely public, and all are welcome (and encouraged) to attend.

Nelson, Pater and Prickett UCLA colloquium

Max Nelson, Joe Pater and Brandon Prickett presented “Representations in neural network learning of phonology” in the UCLA colloquium series Friday October 9th. The abstract is below, and the slides can be found here.

Abstract. The question of what representations are needed for learning of phonological generalizations in neural networks (NNs) was a central issue in the applications of NNs to learning of English past tense morphophonology in Rumelhart and McClelland (1986) and in following work of that era. It can be addressed anew given subsequent developments in NN technology. In this talk we will present computational experiments bearing on three specific questions:

Are variables needed for phonological assimilation and dissimilation?  

Are variables needed to model learning experiments involving reduplication (e.g.  Marcus et al.  1999)?  

What kind of architecture is necessary for the full range of natural language reduplication?  

Vasishth colloquium Friday September 25 at 3:30

Shravan Vasishth (vasishth.github.io), University of Potsdam, will present “Twenty years of retrieval models” in the Linguistics colloquium series at 3:30 Friday September. An abstract follows. All are welcome!

Register here:
https://umass-amherst.zoom.us/meeting/register/tJUldemurz4oGdAo6hV69nh4k3y82zRiLVZB

Abstract

After Newell wrote his 1973 article, “You can’t play twenty questions with nature and win”, several important cognitive architectures emerged for modeling human cognitive processes across a wide range of phenomena. One of these, ACT-R, has played an important role in the study of memory processes in sentence processing.  In this talk, I will talk about some important lessons I have learnt over the last 20 years while trying to evaluate ACT-R based computational models of sentence comprehension. In this connection, I will present some new results from a recent set of sentence processing studies on Eastern Armenian.

Reference:
Shravan Vasishth and Felix Engelmann. Sentence comprehension as a cognitive process: A computational approach. 2021. Cambridge University Press. https://vasishth.github.io/RetrievalModels/

Anderson defends thesis on 6/25

Carolyn Anderson will be defending her dissertation on 6/25 at 1PM. Her dissertation is entitled ‘Shifting the Perspectival Landscape: Methods for Encoding, Identifying, and Selecting Perspectives.’ In her thesis, Carolyn explores formal, computational, and experimental models of perspective representation and processing.

Please join us, virtually, to hear Carolyn present her thesis work! Her dissertation defense will be hosted over Zoom, and we ask that people register for this meeting in advance at the link below. See you there

https://zoom.us/meeting/register/tJIsc-uqqDopHt0bBz5HWn27VHkfMpNyrsOy

Carolyn Anderson, Tessa Patapoutian, and Max Nelson at BAICS

The Bridging AI and Cognitive Science workshop at ICLR 2020 is being held virtually on April 26th. UMass’s Carolyn Anderson and Tessa Patapoutian will be presenting a paper titled “Can Neural Network Models Learn Spatial Perspective from Text Alone?” and Max Nelson will be presenting a paper titled “Learning Hierarchical Syntactic Transformations with Encoder-Decoder Networks.” Full papers, and details on virtually attending, can be found at: https://baicsworkshop.github.io/