[Prev][Next][Index]

CSLI Calendar, 16 November, vol. 5:9


       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
-----------------------------------------------------------------------------
16 November 1989                    Stanford                    Vol. 5, No. 9
_____________________________________________________________________________

    A weekly publication of the Center for the Study of Language and
Information (CSLI), Ventura Hall, Stanford University, Stanford, CA 94305-4115
			     ____________

	  CSLI ACTIVITIES FOR THIS THURSDAY, 16 NOVEMBER 1989

12:00 noon		TINLunch
      Cordura 100	Structure Projection via Natural Language
			Gilles Fauconnier
			Department of Cognitive Science
			University of California, San Diego
			(faucon%cogsci@ucsd.edu)
			Abstract below
			
 2:15 p.m.		CSLI Seminar
      Cordura 100	Models of Rational Agency 8
			Michael Bratman, Martha Pollack, Stan Rosenschein
			(bratman@csli.stanford.edu, 
			pollack@warbucks.ai.sri.com, stan@teleos.com)
			Abstract below

 4:15 p.m.		CSLI Colloquium
      Cordura 100	Prepositions as Verbs: A Case Study in Kenny's Problem
			Arcady Leonidovich Blinov
			Institute of Philosophy
			USSR Academy of Sciences
			Abstract in last week's Calendar
			     ____________

	    CSLI ACTIVITIES FOR THURSDAY, 30 NOVEMBER 1989

12:00 noon		TINLunch
      Cordura 100	Stanford's Policies on Intellectual Property Rights
			Jon Sandelin and David Charron,	Licensing Associates
			Jane McLean, Manager of the Software Distribution
			Center (SDC)
			Office of Technology Licensing (OTL)
			Stanford University
			(sandelin@popserver.stanford.edu)
			Abstract below
			
 2:15 p.m.		CSLI Seminar
      Cordura 100	Models of Rational Agency 9
			Michael Bratman, Martha Pollack, Stan Rosenschein
			(bratman@csli.stanford.edu,
			pollack@warbucks.ai.sri.com, stan@teleos.com)
			Abstract below
			     ____________

			     ANNOUNCEMENT

Because of the Thanksgiving holiday, there will be no Thursday events
and no Calendar on 23 November.
			     ____________

			 THIS WEEK'S TINLUNCH
	      Structure Projection via Natural Language
			  Gilles Fauconnier

Several recent approaches to the study of meaning have called into
question the standard division between semantics and pragmatics and
some have replaced the guiding metaphor of interpreted axiomatic
systems with a "cognitive construction perspective."  Under that view,
linguistic expressions carry context-sensitive instructions for
building up discourse.  Among other things, such instructions set up
domains and links between them; they specify satisfaction conditions
within these domains, and matching conditions across them.  A single
sentence often carries out several such operations at once.  Truth
conditions emerge from the resulting configurations (they are not in
principle directly associated with linguistic forms).

One of several kinds of motivation for the "construction" approach is
the relative ease with which it handles some classical puzzles of
natural logic.  A notable consequence is the unexpected similarity
between micro-reasoning processes in core semantics and
macro-reasoning processes (e.g., analogical reasoning); in both cases,
it seems that structure projection plays a crucial role.
			     ____________

		       THIS WEEK'S CSLI SEMINAR
		     Models of Rational Agency 8
	  Michael Bratman, Martha Pollack, Stan Rosenschein
     Synthesizing Machines that Satisfy Intensional Descriptions
			Stanley J. Rosenschein

There is much current interest in building computer systems that sense
and control complex environments, exhibiting attributes of rational
agency.  Designers have often found it useful to describe such systems
in terms of abstract "mental" entities such as belief, desire, and
intention.  Ultimately, however, artificial agents must be implemented
as concrete computational mechanisms, and it is not obvious precisely
what role intensional descriptions should play in the process of
synthesis.  Approaches based on manipulating run-time data structures
syntactically isomorphic to the intensional metalanguage have not
proven entirely satisfactory.  A promising alternative is found in
synthesis techniques that manipulate symbolic representations of
beliefs and goals at compile time to produce run-time information
structures optimized for real-time update and action selection.  This
talk describes our experience in using techniques of this kind to
develop software for robotic agents.
			     ____________

		 NEXT TINLUNCH, THURSDAY, 30 NOVEMBER
	 Stanford's Policies on Intellectual Property Rights
	       Jon Sandelin, David Charron, Jane McLean

Jon Sandelin will review Stanford's polices with regard to
intellectual property ownership and licensing, the position of OTL
(Office of Technology Licensing) within the university, and OTL's
responsibilities in negotiating intellectual property terms in funding
grants/contracts from industry and in the licensing of technology.
David Charron will describe special considerations in the licensing of
computer software.  Jane McLean will present information about the SDC
(Software Distribution Center).  They plan to leave plenty of time for
questions, and encourage people to present specific situations for
discussion.
			     ____________

	       NEXT CSLI SEMINAR, THURSDAY, 30 NOVEMBER
		     Models of Rational Agency 9
	  Michael Bratman, Martha Pollack, Stan Rosenschein

We will have a general summary discussion of the issues that were
raised in the seminar this quarter.
			     ____________

			SYMBOLIC SYSTEMS FORUM
	 A Survey of Recent AI Research at SRI International
			 C. Raymond Perrault
			  SRI International
		  Friday, 17 November, 3:15, 60-61G
         
Members of SRI's AI Center are engaged in programs of research in
machine vision, natural-language processing, and automated reasoning.
We will survey the main directions of recent work in these areas,
touching on cartographic and medical applications of vision research,
spoken language systems, interfaces integrating natural language and
direct manipulation, and planning and reasoning systems applied to
robotics.  Some demonstrations on videotape will be shown.

The talk will be aimed at undergraduates with no assumed background.
Refreshments will be served.
         		     ____________

		   PHILOSOPHY DEPARTMENT COLLOQUIUM
	   Spacetime, EPR, and the Geometry of Gauge Theory
			    J. B. Kennedy
		       Department of Philosophy
		     (kennedy@csli.stanford.edu)
		  Friday, 17 November, 3:15, 90-92Q

A technical version of the paper is on closed reserve in Tanner Library.
			     ____________

		  LINGUISTICS DEPARTMENT COLLOQUIUM
		  The Metaphorical System of English
			    George Lakoff
		  University of California, Berkeley
		Friday, 17 November, 3:30, Cordura 100

A language contains a conceptual system and a system of expression.
The principles governing the expression of concepts depend on the
nature of the conceptual system, which can differ significantly from
language to language.

Over the past decade, it has been shown that the conceptual system of
English contains a very extensive system of conventional metaphor,
whose primary function is to characterize the structure of abstract
concepts, including such fundamental concepts as time, states, change,
causation, purpose, means, modality, etc.  Each such metaphor is a
mapping from one conceptual domain to another preserving cognitive
topology.  At present, hundreds of such mappings have been discovered
in the English system.  Poetic metaphor, scientific metaphor, and
analogical reasoning also make use of this system.

The talk will survey a number of interesting features of the English
metaphor system, and compare it with what little is known of the
metaphor systems of other languages.  In the course of the discussion
such topics will arise as: the nature of idioms, the role of metaphor
in grammar, the way the lexicon accesses the metaphor system, the
relationship between metaphor and semantic roles, and the relationship
between visual perception and abstract reason.

Happy Hour to follow, in the common (ping-pong) room (Cordura 117).
We will take our speaker out to dinner afterward.  We hope you will be
able to join us.
			     ____________

	    COMMONSENSE AND NONMONOTONIC REASONING SEMINAR

There will be no Commonsense and Nonmonotonic Reasoning Seminar on
Monday, 20 November.
			     ____________

			  PHONOLOGY WORKSHOP
		   Foot Structure and Prosodic Form
			     Armin Mester
		 University of California, Santa Cruz
	     Tuesday, 21 November, 7:30 p.m., Cordura 100

This talk will address questions concerning the contrast between
syllabic and moraic (or quantitative) trochees that emerged in recent
work (Hayes 1987) in response to crosslinguistic dependencies between
quantity-sensitivity and foot type (trochaic or iambic).  The
empirical focus will be on Classical Latin and earlier stages of
German (Old and Middle High German).  Sources of evidence to be
investigated include: (i) linking vowels in Old High German compounds
and verbal paradigms; (ii) a bimoraic prosodic unit playing a pivotal
role in Middle High German verse; (iii) iambic shortening and "brevis
brevians" in Classical Latin poetic meter; (iv) effects of a bimoraic
foot in the system of Latin verbal conjugations.
			     ____________

			    STASS SEMINAR
			   Edward N. Zalta
		      (zalta@csli.stanford.edu)
	       Tuesday, 28 November, 3:15, Cordura 100

I will present a theory of situations that falls out as a subtheory
from my theory of objects.  The following are just some of the
consequences of the theory: situations are partial, partially ordered
by the part-of relationship, persistent, and built up out of states of
affairs; some, but not all, are actual; some are maximal; some maximal
situations are worlds; there is a unique actual world; every actual
situation is a part of the actual world; and no actual situation
supports both a state of affairs and its negation.  After presenting
the theory, I'll trace its path through the choice points of situation
theory, show that one natural way to try to model this theory is by
using nonwellfounded sets, and then compare and contrast the theory
with the views of Perry, Barwise, and Israel.
			     ____________

			   SYNTAX WORKSHOP
			    Chu-Ren Huang
	     Tuesday, 28 November, 7:30 p.m., Cordura 100

No abstract available.