Fred Jelinek Seminar Series

In the fall term 2011, we were honored to host first four speakers of the Fred Jelinek Seminar Series (see the Archive). For the spring term 2012, Ivan A. Sag (Stanford University), Mark Steedman (University of Edinburgh), and Geoffrey Leech (Lancaster University) have confirmed their talks as yet.

Current Lecture

Institute of Formal and Applied Linguistics cordially invites you to the fifth lecture of Fred Jelinek Seminar Series

*FRIDAY*, March 23, 2012

PROF. IVAN A. SAG (Stanford University)
SLUICING AND DIALOGUE

The talk takes place on *FRIDAY* March 23, 2012, at 1:30 p.m. at the Faculty of Mathematics and Physics, Malostranske nam. 25, *3rd* floor, room *S4*. PDF invitation.

Abstract:

[Based in part on joint work with Joanna Nykiel (U. of Silesia).] This talk examines various arguments that have been made by Merchant (2001, 2004, 2008, to appear) against direct-interpretation theories of Sluicing and Bare Argument Ellipsis, e.g. those of Ginzburg and Sag (2000) [GSOO] and Culicover and Jackendoff (2005). We focus on the relevant identity condition, including the inadequacy of e-givenness. Indexical resolution data favors a semantic approach more closely linked to to the discourse context. With more careful examination of GS00's proposal, specifically in relation to the role of salient utterance (SAL-UTT) and the maximal question under discussion (MAX-QUD) supplied by context, Merchant's arguments made against the direct-interpretation approach are seen to lose their force. We also examine data from a number of languages which are problematic for any deletion-based analysis of Sluicing, showing how the direct-interpretation approach avoids these difficulties. Finally, we show how GS00's analysis interacts with Ginzburg's (in press) theory of dialogue to provide an account of `sprouting'.

Institute of Formal and Applied Linguistics cordially invites you to the sixth lecture of Fred Jelinek Seminar Series

Monday, March 26, 2012

PROF. MARK STEEDMAN (University of Edinburgh)
THE STATISTICAL PROBLEM OF LANGUAGE ACQUISITION

The talk takes place on Monday March 26, 2012, at 1:30 p.m. at the Faculty of Mathematics and Physics, Malostranske nam. 25, 4th floor, room S1 (428). PDF invitation.

Abstract:

The talk reports recent work with Tom Kwiatkowski, Sharon Goldwater, and Luke Zettlemoyer on semantic parser induction by machine from a number of corpora pairing sentences with logical forms, including GeoQuery and a corpus consisting of real child-directed utterance from the CHILDES corpus. The problem of semantic parser induction and child language acquisition are both similar to the problem of inducing a grammar and a parsing model from a treebank such as the Penn treebank, except that the trees are unordered logical forms, in which the preterminals are not aligned with words in the target language, and there may be noise and spurious distracting logical forms supported by the context but irrelevant to the utterance. The talk shows that this class of problem can be solved if the child or machine initially parses with the entire space of possibilities that universal grammar allows under the assumptions of the Combinatory Categorial theory of grammar (CCG), and learns a statistical parsing model for that space using EM-related methods such as Variational Bayes learning. This can be done without all-or-none "parameter-setting" or attendant "triggers", and without invoking any "subset principle" of the kind proposed in linguistic theory, provided the system is presented with a representative sample of reasonably short string-meaning pairs from the target language.

About the seminar

The seminar series is founded in recognition of the late Professor Frederick Jelinek, an honorary doctor of Charles University and almost for twenty years a guest Professor of our Faculty.

Professor Dr. Frederick Jelinek (1932-2010), dr.h.c. Charles University in Prague, Julian Smith Professor at JHU, Baltimore, MD, USA, of Czech origin, was an outstanding researcher in Electrical Engineering and Computational Linguistics. His breakthrough ideas have lead to a whole new research paradigm - application of stochastic methods - in the field of automatic speech recognition as well as in natural language processing in general. He held leading positions at Cornell University, IBM T. J. Watson Research Center and Johns Hopkins University, and was a guest professor of Charles University in Prague.

Seminars usually start at 1:30pm every other Monday (taking turns with the regular Seminar on Formal Linguistics), in the room S1 (4th floor) at the MFF UK bulding at Malostranske nam. 25, 11800 Prague 1, Czech Republic.

Archive and links to recordings

Date Invited Lecturer Topic (links to recording if available)
Oct 10, 2011 Joakim Nivre, Uppsala University LOST IN THE WOODS? TRANSITION-BASED DEPENDENCY PARSING WITH NON-PROJECTIVE TREES
Nov 7, 2011 Mirjam Fried, Charles University in Prague HUMAN LANGUAGE AS AN EXERCISE IN CREATIVE RECYCLING: WELCOME TO THE WORLD OF GRAMMATICAL CONSTRUCTIONS
Nov 14, 2011 Dan Flickinger, Stanford University COMBINING SYMBOLIC AND STATISTICAL METHODS IN CORPUS-BASED NLP
Dec 12, 2011 Manfred Stede, University of Potsdam FROM OPINION MINING TO TEXT PARSING: TOWARD THE AUTOMATIC ANALYSIS OF EDITORIALS

Content: Magda Ševčíková. Webmasters: Juraj Šimlovič.
Site is valid XHTML 1.0 and valid CSS. Maintained with TED Notepad replacement and Vim text editor.
2007 © Institute of Formal and Applied Linguistics. All Rights Reserved.

Site navigation: