[Prev][Next][Index][Thread]
CSLI Calendar, 3 June 1998, vol. 13:36
C S L I C A L E N D A R O F P U B L I C E V E N T S
______________________________________________________________________
3 June 1998 Stanford Vol. 13, No. 36
______________________________________________________________________
A weekly publication of the
Center for the Study of Language and Information (CSLI)
Stanford University, Ventura Hall, Stanford, CA 94305-4115
____________
ACTIVITIES DURING 3 JUNE TO 12 JUNE 1998
WEDNESDAY, 3 JUNE
12 noon Semantics Workshop
Margaret Jacks Hall 460:146
Discussion on the Upcoming International Conference on
Pragmatics
Elizabeth Traugott
Linguistics, Stanford
AND
Discussion on Polarity Sensitivity as a Lexical Phenomenon
Michael Israel
Linguistics, UC San Diego
Linguistics, Berkeley
http://www-linguistics.stanford.edu/semgroup/
12 noon What Matters to Me and Why
side chapel of Memorial Church
John Perry
Philosophy and CSLI
4:15pm Seminar on Computational Learning and Adaptation (SCLA)
Cordura 100
Learning to Predict User Operations for Adaptive
Scheduling
Melinda Gervasio
Institute for the Study of Learning and Expertise.
Abstract below
4:15pm EE380: Computer Systems Laboratory Colloquium
Gates B03 (NEC Auditorium)
GPS Chips for Consumer Applications
Greg Turetzky
SiRF
http://www.stanford.edu/class/ee380/schedule.html
4:30pm Engineering Economic Systems Colloquium
Room 380:380Y (Math Corner)
Presentation of 3 Student Projects
http://www.stanford.edu/class/eesor406/ten.html
THURSDAY, 4 JUNE
11:00am CCRMA Hearing Seminar
CCRMA Library
Acoustic Transient Classification by Template Correlation
Tim Edwards
Johns Hopkins
Abstract below
4:00pm Xerox PARC Forum
George Pake Auditorium, Xerox PARC
The Lunar Prospector
Alan Binder
The Lunar Research Institute
http://www.parc.xerox.com/ops/projects/forum
4:00pm 8th annual CS 348B rendering competition
SGI Lab, basement level, Sweet Hall
http://www.graphics.stanford.edu/courses/cs348b-98/
4:15pm Frontiers of Neuroscience Seminars
Munzer Auditorium
Cellular and Circuit Mechanisms of Persistent Neural
Activity
Dr. David W. Tank
Lucent Technologies
Host: Dr. William Newsome
http://www.stanford.edu/dept/nbio/Spring98.html
FRIDAY, 5 JUNE
12:15pm Logic Lunch
Room 380:381TN
Random Bit Strings
Katherine St. John
Santa Clara
http://www-logic.stanford.edu/seminars.html
Abstract below
3:15pm Computer Science Database Seminar
Gates B-12
Ontobroker: Ontology Based Access to the WWW
Rudi Studer
Universitat Karlsruhe (visiting Stanford)
Abstract below
3:15pm Cognitive Seminar
Jordan Hall 420:100
Julie Morrison
http://matia.stanford.edu/html/talks.html
MONDAY, 8 JUNE
12 noon Semantics Workshop
Margaret Jacks Hall 460:146
Verbs of Motion and the Expression of Result
Christine Poulin
Stanford University
Abstract below
4:30pm Stanford Digital Libraries Seminar
Gates B08
The Use of Documents in Real Life (or at least Offices)
Annette Adler
Xerox PARC
Abstract below
WEDNESDAY, 9 JUNE
7:30pm Documentary Film and Video Program
History Corner
Screening of Films by 1st year A.M. students
THURSDAY, 11 JUNE
11:00am CCRMA Hearing Seminar
CCRMA Library
Speech Perception and Learning
The Connection Between Acoustic Processing and Reading
Athanassios Protopapas
Scientific Learning
4:00pm Xerox PARC Forum
George Pake Auditorium, Xerox PARC
Dinosaur Research in the U.C. Museum of Paleontology: The
Real Jurassic Park
Mark Goodwin
U.C. Museum of Paleontology
http://www.parc.xerox.com/ops/projects/forum
7:30pm Sociable Syntax/Socio-Rap session
Margaret Jacks Hall 146
Copula Deletion and other myths of AAVE
Tom Wasow and Ivan Sag
Stanford
Abstract below
FRIDAY, 12 JUNE
12 noon Neurobiology Talk
Fairchild D202
Neuromuscular Function Clarified with Monte Carlo
Modeling
Dr. Miriam M. Salpeter
Department of Neurobiology and Behavior
Cornell University
Host: Dr. U. J. McMahan
http://www.stanford.edu/dept/nbio/June1998.html
SATURDAY, 13 JUNE
2:00pm Documentary Film and Video Program
Cubberly Auditorium
Screening of Thesis Projects by Graduating 2nd year A.M.
students
SUNDAY, 14 JUNE
9:30am Stanford Commencement
Stanford Stadium
Main speech by Ted Koppel, ABC Nightline
no tickets required
____________
SEMINAR ON COMPUTATIONAL LEARNING AND ADAPTATION (SCLA)
on Wednesday, 3 June 1998, 4:15pm to 5:30pm
Cordura 100
http://www-csli.stanford.edu/cll/scla.html
Learning to Predict User Operations for Adaptive Scheduling
Melinda T. Gervasio
Institute for the Study of Learning and Expertise
Palo Alto, California
mailto:gervasio@isle.org
http://www.isle.org/~gervasio/crisis.html
Mixed-initiative systems present the challenge of finding an effective
level of interaction between humans and computers. Machine learning
presents a promising approach to this problem in the form of systems
that automatically adapt their behavior to accommodate different
users. This talk presents an empirical study of learning user models
in an adaptive assistant for crisis scheduling. I will begin by
describing HAZMAT, our synthetic problem domain for hazardous
materials incidents, and then INCA, our computational assistant for
crisis response. The results of a baseline study show some benefit
from learning but leave room for improvement. I will also discuss
three subsequent experiments that investigate the effects of problem
reformulation on performance. The results reveal that problem
reformulation leads to significantly better accuracy without
sacrificing the usefulness of the learned behavior. The study also
raises several interesting issues in adaptive assistance for
scheduling.
____________
CCRMA HEARING SEMINAR
on Thursday, 4 June 1998, 11:00am
CCRMA Library
http://www-ccrma.stanford.edu/CCRMA/Events/Events.html
Acoustic Transient Classification by Template Correlation
Tim Edwards
Johns Hopkins
Acoustic transients are a rich source of information in the natural
world. Biological systems process them quickly and economically. We
have developed a biologically-inspired analog VLSI architecture for
real-time processing of acoustic transients. Judicious front-end
manipulation of the signal in the time-frequency domain allows an
elegant and robust implementation of a template correlation algorithm.
The algorithm reduces continuous-valued template memory and
multiplication to simple multiplexing and binary storage, yet retains
acceptable performance on a typical classification task in simulation.
Measurements from the analog VLSI processor confirm that, as is true
of neural systems in general, massively parallel collective
computation yields accurate, reliable results from inaccurate,
unreliable components.
____________
LOGIC LUNCH
on Friday, 5 June 1998, 12:15pm
Math Corner 380:381T
http://www-logic.stanford.edu/seminars.html
Random Bit Strings
Katherine St. John
Santa Clara University
The ordered structures with a single unary predicate -- that is, bit
strings -- are a natural class to study both for their beauty as
mathematical structures and their role in computer science. Much work
has been done linking classes of bit strings and ordered graphs to
complexity classes. In this talk, we will discuss properties of
random graphs and bit strings, in particular, joint work with Joel
Spencer on sparse random bit strings.
The random bit string, U_{n,p}, is a probability
space over unary predicates U on [n] = {1,...,n} with the
probabilities determined by
Pr[U(x)] = p(n), for 0 < x < n+1,
and the events U(x) are mutually independent over
0 < x < n+1.
We are interested in what happens to properties of the random bit
string as the length of the string goes to infinity. How does this
change as the underlying probability p(n) changes? Also, what do the
models of the sentences that are almost surely true (e.g., the almost
sure theory) look like? We will give an overview of this work and
discuss what happens at some interesting probabilities.
____________
COMPUTER SCIENCE DATABASE SEMINAR
on Friday, 5 June 1998, 3:15pm - 4:30pm
Gates B12
http://www-db.stanford.edu/dbseminar/dbseminar.html
Ontobroker: Ontology Based Access to the WWW
Rudi Studer
Institute AIFB, University of Karlsruhe, Germany
on sabbatical at the InfoLab, Stanford University
The World Wide Web (WWW) is currently one of the most important
electronic information sources. However, its query functionality and
the provided reasoning services are rather limited.
Ontobroker consists of a number of languages and tools that enhance
query access and inference service in the WWW. It provides languages
to annotate Web documents with ontological information, to represent
ontologies and to formulate queries based on these ontologies.
Ontobroker allows to access directly the Web information you are
interested in and to infer new knowledge with an inference engine
based on techniques from logic programming. This talk uses examples
from the 'Knowledge Annotation Initiative of the Knowledge Acquisition
Community' to illustrate these languages and tools and the kind of
services that are provided.
The Ontobroker system is available on the Web at:
http://www.aifb.uni-karlsruhe.de/WBS/broker/
The Ontobroker system was developed jointly with Stefan Decker,
Michael Erdmann and Dieter Fensel.
Biography: From January 1977 to June 1985 Rudi Studer worked as a
research scientist at the University of Stuttgart. From July 1985 till
October 1989 he was project leader and manager at IBM Germany
Scientific Centre, Institute of Knowledge Based Systems. Since
November 1989 he has been full professor in Applied Computer Science
at the University of Karlsruhe. He received his Doctor's degree in
Mathematics and Computer Science from the University of Stuttgart in
1982. His research interests include Knowledge Engineering, Knowledge
Discovery in Databases and Knowledge Management.
____________
STANFORD SEMANTICS WORKSHOP
on Monday, 8 June 1998, 12:00pm
Margaret Jacks Hall 460:146
http://www-linguistics.stanford.edu/semgroup/
Verbs of Motion and the Expression of Result
Dissertation Proposal
Christine Poulin
In his ground-breaking work on lexicalization patterns, Talmy (1985)
presents a typology of motion verbs based on the category of
information that conflates with motion in the main verb, e.g., Path or
Manner. He finds that in one type of language, such as English, the
main verb typically encodes manner along with the fact of motion,
while the path (i.e., the general direction followed relative to the
ground) is expressed by elements associated with the verb ("I ran up
to the gym"). In contrast, in another type of language, such as
French, the main verb encodes the path along with the fact of motion,
and the manner is expressed by other elements such as gerundives ("Je
suis monte au gym en courant"). While Talmy talked mostly about
tendencies, subsequent work has focussed on claims about
well-formedness. For example, Rappaport Hovav and Levin (1995) have
proposed that French-like languages can only use manner-of-motion
verbs to express motion within a location ("J'ai couru au gym" to mean
only `I ran at the gym', not `I ran to the gym'). Slobin and Hoiting
by contrast claim that French-like languages can express change of
location, but only if no boundaries are crossed ("J'ai couru jusqu'au
gym" means `I ran to the gym'; but "J'ai couru dans le gym" cannot
mean `I ran into the gym').
In this talk (based in part on earlier work with Vivienne Fong) I try
to sort out what is and is not grammatical in French, drawing on data
from many native speakers as well as a large corpus of written
sentences. I present evidence that the relevant syntactic
constructions in French and English are essentially identical. Where
they differ is in the types of verbs that are associated with the
different semantic templates, which results in different syntactic
subcategorization. In particular, English freely allows its verbs of
motion to undergo a template expansion that adds a BECOME (change of
state) predicate to the basic Activity template. This option is not
available to French, and so it can express "motion to" only if the
verb or the preposition in question comes pre-equipped with a BECOME
predicate. Therefore in French, sentences like "J'ai couru dans le
gym" are attested in the meaning `I ran into the gym', because the
verb is pre-equipped with a BECOME predicate; "J'ai nag vers la rive
droite" can mean `I swam towards the right bank' because the
preposition itself includes the BECOME predicate; but "J'ai nag dans
la caverne" can mean only `I swam within the cave', because no lexeme
has the BECOME predicate. The analogous English sentence, `I swam in
the cave', can mean either `within' or `into', because of the
availability of template expansion within English. Among the evidence
that English verbs can undergo change-of-state template expansion is
that the syntactic structure that can be used for depictives ("They
burned the witch alive") can also be used for resultatives ("They
burned the witch dead"), which extension is unavailable in French. I
will present results from a pilot study that indicates that this
connection between the behavior of motion verbs and resultatives is
not accidental but holds up cross-linguistically. I will also seek
feedback for a proposal to investigate the semantic factors that
determine which manner-of-motion verbs are associated with BECOME
predicates in French-like languages; and ideas for a historical
investigation of how Latin changed from an English-like language
(i.e., with template expansion) to a French-like language.
References:
Rappaport Hovav, Malka and Beth Levin. 1995. Morphology and lexical
semantics. To appear in A. Zwicky and A. Spencer (eds), Handbook of
morphology. Oxford: Blackwell.
Slobin, Dan and Nini Hoiting. 1994. Reference to movement in spoken
and signed languages: typological considerations In Proceedings of the
Annual Meeting of the Berkeley. Linguistic Society, 20, 487-505.
Talmy, Leonard. 1985. Lexicalisation patterns: Semantic structure
in lexical forms. In T. Shopen (ed.), Language typology and syntactic
description, Vol. 3: Grammatical categories and the lexicon,
57-149. Cambridge: Cambridge University Press.
____________
STANFORD DIGITAL LIBRARIES SEMINAR
on Monday, 8 June 1998, 4:30pm
Gates Building, B08
http://diglib.stanford.edu/diglib/seminars/seminars.html
The Use of Documents in Real Life (or at least Offices)
Annette Adler
Xerox PARC
This talk is about how documents get used. It describes a project,
Goldfish, that I've been involved with for two years. Goldfish
considers how people in a variety of office settings use documents,
emerging document types (beyond just static digital, e.g. audio) and
document appliances (such as pagers and pilots), particularly in the
context of:
- intermingled virtual and physical environments (and how folks
negotiate getting back and forth)
- heterogeneous media (digital, paper, voice and other forms)
- blended personal/workgroup/institutional/public foci
- multiple locations for work, both predictable and unpredictable
- black and white as well as color documents
One particularly interested type of document use is reading. This
talk will describe a study focusing on this aspect in some detail.
About the speaker: I am a researcher at Xerox with a formal background
in social anthropology and a working background in systems
architecture. Integrating these two has lead to many interesting and
fruitful tensions. My interests include understanding use (of systems
by people and systems themselves), mediated collaboration and
community and chipping away at the divides between virtual/physical
worlds.
____________
SOCIABLE SYNTAX/SOCIO-RAP SESSION
on Thursday, 11 June 1998, 7:30pm
Margaret Jacks Hall 460:146
Copula Deletion and other myths of AAVE
Tom Wasow and Ivan Sag
Stanford
One evident difference between African American Vernacular English
(AAVE) and Standard American English (SAE) in that the latter requires
finite verbs in certain environments where the former does not. The
clearest case of this is the so-called 'copula deletion' construction,
illustrated by (1):
(1)a. Chris at home
b. We angry with you.
c. You a genius!
d. They askin for help.
Other examples include what have been termed 'aspect markers' such as
invariant 'be', BIN, and 'done', as well as the absence of the
third-person singular -s suffix on verbs:
(2)a. I be eatin then.
b. The children BIN ate.
c. The children done ate.
d. Andy make a lot of money.
We argue that (1) can be handled straightforwardly in HPSG as a
headless clause, without recourse to deletion rules or phonetically
null elements. Such an analysis brings out the commonalities between
verbless clauses in AAVE and those in other languages such as Russian
and Hungarian.
We analyze the examples in (2a-c) as headed by finite, non-auxiliary
verbs. (2d) illustrates a morphological fact about AAVE, reflecting
the near disappearance of subject-verb agreement from this language.
But we show that all of the examples in (2) can be treated as finite
clauses, providing a straightforward account of the syntax and
semantics of such sentences.
Finally, we consider Labov's case for deletion account of examples
like (1). In particular, we go in detail through the list of
environments that he claims are exceptions to copula deletion, showing
that our analysis gives a consistently more elegant and less
stipulative account of the phenomena.
____________
END MATERIAL
The CSLI Calendar appears weekly on Wednesdays throughout the academic
year. Announcements, abstracts, and other information to appear in
the Calendar should be submitted to the editor, who reserves the right
to decide what does or does not go in the calendar
mailto:incalendar@csli.stanford.edu.
Requests to be added to the mailing list should be sent to
mailto:majordomo@csli.stanford.edu. With the lines in the body of
the text of either
subscribe csli-calendar
for the long form or
subscribe csli-short-calendar
for the short form. Problems with subscribing or unsubscribing should
be sent to owner-csli-calendar@csli.stanford.edu.
The full current issue is at
http://www-csli.stanford.edu/csli/Archive/calendar/current.html
and the archives at
http://www-csli.stanford.edu/csli/Archive/calendar/.
People on most of the CSLI computers can type 'help csli-calendar' to
see the current issue.
The CSLI Calendar is also posted each week to
news://nntp-csli.stanford.edu/csli.bboard.
Information about CSLI's research program is available at
http://www-csli.stanford.edu/csli/.
For maps to the Stanford University campus see
http://www.stanford.edu/home/visitors/maps.html.
____________