[Prev][Next][Index]
CSLI Calendar, 30 MAY 1996, vol.11:29
-
To: friends@Turing.Stanford.EDU
-
Subject: CSLI Calendar, 30 MAY 1996, vol.11:29
-
From: Trudy Vizmanos <trudy@csli.Stanford.EDU>
-
Date: Wed, 29 May 1996 15:03:55 -0700 (PDT)
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
______________________________________________________________________________
30 May 1996 Stanford Vol. 11, No. 29
______________________________________________________________________________
A weekly publication of the Center for the Study of Language and
Information (CSLI), Stanford University, Ventura Hall, Stanford, CA 94305-4115
____________
CSLI ACTIVITIES DURING 30 MAY -- 7 JUNE 1996
THURSDAY, 30 MAY
11:00 - Special Talk
Ventura Hall, Room 17
Metaphor and Lexical Semantics
James Pustejovsky, Brandeis Computer Science
3:15 - Philosophy Department Colloquium
Encina Hall, Room 423
Situated Objectivity
Robert Kraut, Ohio State University
and Stanford Humanities Center
4:15 - SSP Forum
Building 60, Room 61F
Symbolic Systems Honors Projects, 1996
Jith Meganathan and Kevin Henry
7:30 - Stanford Phonology Workshop
Margaret Jacks Hall, Room 146
Boundary Tones and the Phonetic Implementation of Tone
in Chichewa
Scott Myers, University of Texas
FRIDAY, 31 MAY
12:30 - HCI Seminar
Skilling Auditorium
The interactions design awards
Lauralee Alben (Alben+Ferris), Harry Saddler (Apple),
Terry Winograd (Stanford) [mail@albenfaris.com]
Abstract below
MONDAY, 3 JUNE
1:00 - Linguistics Talk I
Margaret Jacks Hall, Room 146
On the Comprehension/Production Dilemma in Child Language
Paul Smolensky, Johns Hopkins University
Abstract below
4:30 - Linguistics Talk II
Margaret Jacks Hall, Room 146
On the Explanatory Adequacy of Optimality Theory
Paul Smolensky, Johns Hopkins University
Abstract below
TUESDAY, 4 JUNE
6:30 - SSP Film Series
Cubberley Education Building
The Man Who Believed in Body Transplants (1989)
and To Be (1991)
Abstract below
____________
The CSLI Calendar appears weekly on Wednesdays throughout the academic year.
Announcements, abstracts, and other information to appear in the Calendar can
be submitted to [mailto:incalendar@csli.stanford.edu].
Information about CSLI's research program and past issues of the CSLI Calendar
are available at [http://www-csli.stanford.edu/csli/]. The CSLI Calendar is
also posted each week to [news://nntp-csli.stanford.edu/csli.bboard].
____________
FIFTH CSLI WORKSHOP ON LOGIC, LANGUAGE, AND COMPUTATION
on 31 May - 2 June
Cordura Hall, Room 100
This annual event brings together philosophers, linguists, and computer
scientists with an interest in logic, with the overall aim of facilitating
interdisciplinary interaction. The previous four installments have been
pleasant and productive, with a mix of participants from (mainly) California
and The Netherlands.
The Workshop is organized by Johan van Benthem, Henriette de Swart, Rob van
Glabbeek, and Jean Braithwaite
For information, contact (braith@csli.stanford.edu).
Webpage http://www-csli.stanford.edu/users/kyle/llc5.html
WORKSHOP PROGRAM
Each talk will consist of 30 minutes of presentation followed by a fifteen
minute question and discussion period.
FRIDAY, MAY 31: COMPUTATION CHAIR: GRIGORI MINTS
9:00-9:15 Opening Remarks
9:15-10:00 ALBERT VISSER
Two faces of dynamic interpretation: Relations and
partial information states
10:00-10:45 GERARD RENARDEL
A Variant of Quantified Dynamic Logic
10:45-11:00 Coffee Break
11:00-11:45 ATOCHA ALISEDA
Toward a Logic of Abduction
11:45-12:30 NIR FRIEDMAN
Plausibility Measures and Default Reasoning
12:30-1:30 Lunch Break
II. LANGUAGE AND COMPUTATION CHAIR: LIVIA POLANYI
1:30-2:15 JAMES PUSTEJOVSKI
The semantics of complex types
2:15-3:00 VIJAY SARASWAT
Linear concurrent constraint programming as a basis
for semantic interpretation in LFG
3:00-3:15 Coffee Break
3:15-4:00 WILLEM GROENEVELD
Dynamic Epistemic Logic
4:00-4:45 MARK GAWRON
Questions and the Semantics of the English Universal
Concessive Conditional
SATURDAY, JUNE 1
III. LANGUAGE CHAIR: TOM WASOW
9:15-10:00 JACK HOEKSEMA
Systems of negative concord
10:00-10:45 DONKA FARKAS
Distributed Indefinites
10:45-11:00 Coffee Break
11:00-11:45 YOOKYUNG KIM
A Situation Semantic Account of Existential Sentences
11:45-12:30 FRANS ZWARTS
Determinants of scope and negation in the language of children
and adults
12:30-1:30 Lunch Break
IV. LANGUAGE AND LOGIC CHAIR: ED ZALTA
1:30-2:15 JAAP VAN DER DOES
Interpreting Nominal Anaphora by means of Scope Extension
2:15-3:00 MANFRED KRIFKA
Frameworks for the Representation of Focus
3:00-3:15 Coffee Break
3:15-4:00 JAN-TORE LOENNING
Plural quantification, predication, and ontology
4:00-4:45 MARTIN STOKHOF
Objects and Concepts? Perspectives in multi-speaker discourse
SUNDAY, JUNE 2
V. LOGIC CHAIR: STEVE GIVANT
9:15-10:00 ANTONIA HUERTAS & MARIA MANZANO
Partial and heterogeneous tools for building new logics
10:00-10:45 MICHIEL VAN LAMBALGEN
Local quantification, or poor man's probability
10:45-11:00 Coffee Break
11:00-11:45 MARTIN GROHE
The Complexity of Logical Equivalence and Bisimulation
11:45-12:30 KIT FINE
Semantics for the Logic of Essence
12:30-1:30 Lunch Break
VI. LOGIC AND COMPUTATION CHAIR: JOSE MESEGUER
1:30-2:15 NATARAJAN SHANKAR
Model Checking and Theorem Proving in PVS
2:15-3:00 HENNY SIPMA
Deductive Modelchecking
3:00-3:15 Coffee Break
3:15-4:00 TOM COSTELLO
Limit Circumscription
4:00-4:45 VAUGHAN PRATT
An Abstract Notion of Language
____________
INTENSIONAL LOGIC
1:15 p.m., Room 550D
As usual, the final part of this course is devoted to presentations of special
topics from current research.
May 13: Modal Remodeling for Predicate Logic 1
(how to make first-order logic decidable by generalizing Tarski
semantics)
JOHAN VAN BENTHEM
May 15: Modal Remodeling for Predicate Logic 2
(general consequences of this viewpoint)
JOHAN VAN BENTHEM
May 20: Dynamic Epistemic Logic
(many-person updates in epistemic logic)
Willem Groeneveld (Amsterdam)
May 22: Modal Logic, Representation and Translation
(analyzing modal logics via other, less or more, multi-purpose
standard logics)
MARIA MANZANO (Barcelona)
May 29: A Philosophical Conception of Modal Logic
(a view of modal logic leading to strong intensional theories with
both historic
(Leibniz, Frege) and systematical uses)
ED ZALTA (Stanford)
LAST WEEK: There may be presentations by some Dutch visitors to the `5th CSLI
Workshop in Logic, Language and Computation', which will be announced
separately.
To get further information concerning course contents, as well as reading
materials, please contact {johan, willem, manzano, zalta} @csli.stanford.edu,
respectively.
____________
SPECIAL TALK
on Thursday, 30 May
11:00 a.m., Ventura Hall, Room 17
Metaphor and Lexical Semantics
James Pustejovsky
Brandeis Computer Science
____________
SSP FORUM
on Thursday, 30 May
4:15 p.m., Building 60, Room 61F
Symbolic Systems Honors Projects, 1996
Jith Meganathan and Kevin Henry
(Seniors in Symbolic Systems)
JITH MEGANATHAN: My honors thesis, which is somewhat outside the
traditional domain of Symbolic Systems, is an outgrowth of research I
performed with Dr. Roy King in the Department of Psychiatry. This purpose
of this research was to investigate the processes underlying psychiatric
art therapy, in which a group of patients draws or paints under the
guidance of a therapist; in particular, the study focused on the
relationships between patients' preferences for visual complexity and the
affective nature of their art. My primary responsibility was to develop
methods for evaluating the emotional expressiveness of patient art. My
presentation will focus on the methods I developed and their relationship
to the larger hypotheses governing the study. I will conclude by
presenting our results and their connections to the study of the mind.
KEVIN HENRY: Go is one of the few board games which computers have
yet to master. Chess is a game dominated by tactical considerations, and
computers have achieved success by using brute-force search techniques.
Backgammon is dominated by judgemental considerations, for neither humans
nor computers can search very deeply due to the large number of possible
moves per turn and the stochastic effects of the dice. Instead, the best
backgammon program, which has achieved world-class performance, is a
neural network which has learned to estimate the value of a backgammon
position.
Go requires a great deal of both strategic and tactical prowess, and neither
brute-force search nor purely judgemental techniques have proven
effective. This project is an initial attempt to combine the tactical strength
of search procedures with the judemental ability of neural
networks. Specifically, a neural network is given as input an estimate of the
"life" of each stone on the board, and is trained on a small sample of
professional Go games to choose the best move. This input value, which
represents the key concept in Go, is calculated by performing short, local
searches. This may ultimately prove to be an effective approach to the
problem.
(A third Symbolic Systems honors student from this year, Marc Pauly,
finished his thesis in March on "Godel's Second Incompleteness Theorem."
Marc is away this quarter and will be unable to give a presentation.)
____________
STANFORD PHONOLOGY WORKSHOP
on Thursday, 30 May
7:30 p.m., Margaret Jacks Hall, Room 146
Boundary Tones and the Phonetic Implementation
of Tone in Chichewa
Scott Myers
University of Texas
____________
SEMINAR ON HUMAN-COMPUTER INTERACTION
on Friday, 31 May
12:30 p.m., Skilling Auditorium
The Interactions Design Awards
Lauralee Alben (Alben+Ferris), Harry Saddler (Apple),
Terry Winograd (Stanford) [mail@albenfaris.com]
In the Spring of 1995, a new design competition was announced by
interactions magazine. The call for entries stated:
------
The practice of interaction design has moved quickly to the foreground in
the software field. And, as more and more things in the world become
interactive, interaction design is becoming an important aspect in many
other fields. Despite all this growth, we are all just beginning to learn
how to do this. Designing interactive experiences is so danged hard that
all of us in the field have a need to learn from the successes of others.
To help this learning process along, the Association for Computing
Machinery (ACM), through interactions, its quarterly magazine on design, is
developing an annual recognition program for people who design interactive
products. This peer-recognition program will set a high standard for the
industry, not only recognizing achievement but encouraging designers and
companies to reach for quality. The criteria for judging, the credentials
of the jurors, and the nature of the award process will combine to make the
interactions award a prestigious program. The interactions award is
modeled on programs sponsored by publishers in other design-oriented
fields, such as Communication Arts' Design Annuals and the IDEA awards for
industrial design....
This isn't an award for good looks or clever innovations. We aren't going
to recognize the fanciest new widget. We aren't looking for the
best-selling, the most adrenaline-inducing, or the best use of video. Not
that these things are bad--we like an adrenaline-inducing multimedia
best-seller with clever widgets as much as the next techno-sapien. But when
it comes to awards, we care about quality of interaction. We want to
recognize products, services, and environments that enhance people's lives.
We're looking for designs that effectively help people work, learn, live,
play or communicate.
------
The results of the competion were presented in the May+June, 1996 issue of
interactions. They were the product of a review process that was deeply
concerned with figuring out what criteria should be applied to interaction
design, as well as judging the individual entrants. The primary issue was
Quality of experience Taken together, the criteria raise one key question:
How does effective interaction design provide people with a successful and
satisfying experience? The detailed criteria that evolved were:
*Needed
*Understanding of users
*Learnable,
*Usable
*Appropriate
*Effective design process
*Aesthetic experience
*Manageable
*Mutable
These are described more fully in the magazine (excerpted on-line in the
web pages for this course). We will show a video that was made about the
review process, then have comments and discussion by three of the people
who participated. The focus will be on our struggle to define what is meant
by "quality" in interaction design, and on examples from the competition
that illustrate the best features of interactive hardware and software.
____________
PHILOSOPHY DEPARTMENT COLLOQUIUM
on Friday, 31 May
3:15 p.m., Encina Hall, Room 423
Situated Objectivity
Robert Kraut
Ohio State University & Stanford Humanities Center
____________
SEMANTICS WORKSHOP
on Monday, June 3
11:00 a.m., Corudra Hall, Room 104
'Only', Non-Monotonicity, and Negative Polarity Items
Jay Atlas
Pomona College
Work by Ladusaw, Horn, and others suggests that `Only Gs F'
and `Only a Fs' should be analyzed as having downwards monotonic
'Only X' quantifiers of the same semantic type. To the contrary,
I show that `Only a' is not downwards montonic, but is what in
the typology of Zwarts (1996) and Atlas (1995) I call a
Pseudo-Regular negative quantifier. I discuss the adequacy of
Horn's arguments in support of the downwards monotonic view,
and show why they are defective. In consequence there arises
a number of new questions about the taxonomy of Negative
Polarity Items. I show why the analysis of Atlas (1991, 1993)
for `Only a Fs' successfully explains the data.
____________
LINGUISTICS TALK I
on Monday, 3 June
1:00 p.m., Margaret Jacks Hall, Room 146
On the Comprehension/Production Dilemma in Child Language
Paul Smolensky
Johns Hopkins University
In this short (squib-length) paper I consider a natural extension of
Optimality Theory, developing a competence theory in which
`comprehension' and `production' are formalized as two related
functions defined by two kinds of opimization with a single
grammar/ranking. I argue that this slightly extended OT can resolve a
longstanding dilemma in generative theories of language acquisition. A
single grammar of the sort being proposed in current OT acquisition
work is shown to allow comprehension which is rich in distinctions
which are neutralized in production.
____________
LINGUISTICS TALK II
on Monday, 3 June
4:30 p.m., Margaret Jacks Hall, Room 146
On the Explanatory Adequacy of Optimality Theory
Paul Smolensky
Johns Hopkins University
Originally centered in the context of phonology, discussion of
the explanatory adequacy of OT has mainly concerned the explicit or
implicit comparison with grammars based entirely or heavily on ordered
rules. In the past few years, work on OT syntax has brought to the
fore explanatory issues arising instead from the comparison with other
grammatical theories based on contraints: inviolable ones. This
paper identifies 14 claims against the explanatory adequacy of
OT, against the backdrop of inviolable-constraint-based grammatical
theories in syntax and phonology.
In all, 14 challenges to the explanatory power of OT are examined.
Then, doubt is put aside and the 15th point highlights two particularly
interesting explanatory strategies that have proved powerful within OT
(syntax as well as phonology).
I. `Output' Constraints, Derivation, and Filtering Gen
(1) "OT says that the grammatical forms are those that optimally
meet output constraints, so every word should surface as
ba [sic]"
(2) "Opacity effects (like Hebrew spirantization) are not
naturally accounted for in (parallel) OT and are not
treated in the OT literature."
(3) "OT is an inadequate theory because syntax/phonology is
derivational, and OT isn't."
(4) `Since Gen typically generates an infinite number of
candidates, OT is psychologically impossible.'
II. OT Constraints and Explanation
(5) `OT is unexplanatory because you can make up any constraint
you want; in OT, you just give up on seeking explanatory
factors.'
(6) `Any framework which leads to the morass of constraints
found in OT analyses in phonology cannot possibly be
explanatorily adequate.'
III. Cross-linguistic Variation and Constraint Re-ranking
(7) `Constraint ranking is as unexplanatory as extrinsic rule
ordering.'
(8) `Re-ranking is a totally unconstrained theory of cross-linguistic
variation.' Relative to: principles and parameters.
(9) `Re-ranking is essentially the same as parameters which turn
on/off any constraint: burying the constraint at the
bottom of the hierarchy turns it off, and putting it at
the top of the hierarchy makes it inviolable.'
(10) `The lexicon is a better locus for cross-linguistic
variation because it has to be learned anyway.'
(11) `The goal is a perfect syntax ... of course, the lexicon,
phonology, morphology are not going to be perfect.'...
I.e.: *Complicate-Syntax >> *Complicate-Lexicon/Phonology/Morphology
(12) `Ranked violable constraints can always be replaced by
(unranked) inviolable constraints.'
IV. Re-ranking, Restrictiveness of UG, and Learnability
(13) `Re-ranking gives wildly too many grammars to be
explanatory.'
(14) `Re-ranking gives wildly too many grammars to be learnable.'
V. Beyond Doubt
(15) Two central explanation patterns of OT
a. OT Idea: (P&S; McCarthy & Prince 93: 'emergence of the
unmarked')
The Subordination Pattern of explanation:
A constraint which is undominated and unviolated
in language L1, is dominated in another
language L2, where it is violated, but
still active, emerging in environments in
which no dominant constraint contravenes.
b. OT Idea: (P&S)
Markedness scales formalized as Universal
Constraint Sub-Hierarchies:
Roughly: a markedness hierarchy
A *> B *> C
('*>' = 'is more marked than') is formalized
as the universal constraint sub-hierarchy
*A >> *B >> *C.
I.e., UG imposes this sub-hierarchy, so that
every allowed grammar ranks these constraints
in this sequence; other constraints may
interrupt the sub-hierarchy.
____________
SSP FILM SERIES
on Tuesday, 4 June
6:30 p.m., Cubberley Education Building
The Man Who Believed in Body Transplants (1989)
and
To Be (1991)
Our last videos this year raise questions about philosophy, personal
identity, and scientific ethics. "The Man Who Believed in Body
Transplants" is a half-hour BBC production featuring interviews with
neurologist Robert Joseph White on how he justifies the nature of his
brain transplantation research. In the early 1970s, Dr. White carried out
very controversial experiments in which he disembodied and sometimes
transplanted the heads and brains of monkeys. The second video ("To Be")
is a 10-minute animated short from the National Film Board of Canada.
The Symbolic Systems Film Series showcases films and tapes of general
cognitive science interest. Attendance at film series events can
substitute for attendance at the Symbolic Systems Forum for students
enrolled in SSP 10 for one unit. All are welcome at these events. The
showing of the videos is followed by a discussion, and researchers who
are knowledgeable about the program's topic are urged to join us in
evaluating it.
____________