[Prev][Next][Index]
CSLI Calendar, 4 May 1995, vol.10:26
-
To: friends
-
Subject: CSLI Calendar, 4 May 1995, vol.10:26
-
From: Tom Burke <burke>
-
Date: Wed, 3 May 1995 12:52:22 -0700
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
______________________________________________________________________________
4 May 1995 Stanford Vol. 10, No. 26
______________________________________________________________________________
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 3 -- 12 MAY 1995
WEDNESDAY, 3 MAY
4:00 - Seminar on Computational Learning and Adaptation
Cordura Hall, Room 100
Finding Propositions from Numerical Data and Neural Networks
using Multi-Linear Functions
Hiroshi Tsukimoto, Toshiba Corporation
Abstract below
7:00 - Philosophy Talk
Building 90, Tanner Library
A Discussion about Deconstruction
Rene' Girard, Stanford French and Italian
Abstract below
THURSDAY, 4 MAY
10:00 - STASS Seminar
Cordura Hall, Room 100
Interpretive Grammar and Stylized Documents
Keith Devlin, CSLI & St. Mary's College
Abstract below
12:15 - CSLI Seminar
Series on Intelligent Agents
Cordura Hall, Room 100
Perception, Action, and Cognition in an Integrated
Architecture for Physical Agents
Pat Langley, Stanford Robotics Laboratory
Abstract below
4:15 - SSP Forum
Building 60, Room 61-F
The Virtues of Lightness
Johan van Benthem, Amsterdam and Stanford Philosophy
Abstract below
FRIDAY, 5 MAY
12:00 - Logic Lunch
Building 380, Room 383-N
Parametricity and Variants of Girard's J Operator
John Mitchell, Stanford Computer Science
Abstract below
12:30 - HCI Seminar
Skilling Auditorium
The Future of the Internet
Marc Andreesen, Netscape Communications Corporation
Abstract below
3:15 - Philosophy Colloquium
Building 90, Room 91-A
Physicalism and Physics
Yair Guttmann, Stanford Philosophy
3:30 - Linguistics Colloquium
Cordura Hall, Room 100
Grammatical Weight
Tom Wasow, Stanford Linguistics
Abstract below
MONDAY, 8 MAY
1:15 - Intensional Logic Lecture
Building 90, Room 92-Q
Modal Phrase Structure Grammar
Johan van Benthem, Amsterdam and Stanford Philosophy
WEDNESDAY, 10 MAY
1:15 - Intensional Logic Lecture
Building 90, Room 92-Q
Relevant Logic
Phil Kremer, Stanford Philosophy
THURSDAY, 11 MAY
12:15 - CSLI Seminar
Series on Intelligent Agents
Cordura Hall, Room 100
Toward Building an Android: Recent Developments
of the Cog Project
Cynthia Ferrell, MIT AI Laboratory
Abstract below
4:15 - SSP Forum
Building 60, Room 61-F
Recent Developments of the Cog Project (cont'd)
Cynthia Ferrell, MIT AI Laboratory
Abstract below
FRIDAY, 12 MAY
12:30 - HCI Seminar
Skilling Auditorium
The Juno-2 Constraint-Based Drawing Editor
Allan Heydon, DEC Research
3:15 - Philosophy Colloquium
Building 90, Room 91-A
Title to be announced
Marcia Cavell
3:30 - Linguistics Colloquium
Cordura Hall, Room 100
Title to be announced
Jaye Padgett, UC Santa Cruz
____________
The CSLI Calendar appears on Thursday of each week throughout the academic
year. Announcements, abstracts, and other information to appear in the
Calendar on a given Thursday should be submitted by e-mail to
<incalendar@csli.stanford.edu> by 5:00 p.m. on the previous Tuesday.
Past issues of the CSLI Calendar, a quarterly schedule of upcoming CSLI
events, and other information about CSLI are available at URL
<http://www-csli.stanford.edu/>. The Calendar, with available abstracts, is
also posted each week to the csli.bboard newsgroup.
____________
SEMINAR ON COMPUTATIONAL LEARNING AND ADAPTATION
on Wednesday, 3 May
4:00 p.m., Cordura Hall, Room 100
Finding Propositions from Numerical Data and
Neural Networks using Multi-Linear Functions
Hiroshi Tsukimoto
Toshiba Corporation
A method for finding propositions from numerical data and neural networks is
presented. The method is based on the space of multi-linear functions, an
extension of Boolean algebra that can be made into a Euclidean space when the
domain of the functions is {0,1} or [0,1]. Using a method such as multiple
regression analysis, numerical data can be approximated by (multi-)linear
functions, which can then be approximated by (extended) Boolean functions.
Therefore, propositions can be obtained from numerical data. When the output
function is a sigmoid function, the units of neural networks after learning
can be approximated by (multi-)linear functions, which can be approximated by
(extended) Boolean functions. Therefore, propositions can be obtained from
neural networks. A few examples of this approach are presented.
The goal of this seminar is to increase communication among local researchers
with interests in computational approaches to learning and adaptation. If you
would like to be added to (or removed from) the mailing list, or if you are
interested in giving a talk in the seminar, please send email to
<langley@cs.stanford.edu>.
____________
PHILOSOPHY TALK
on Wednesday, 3 May
3:15 p.m., Building 90, Room 91-A
A Discussion about Deconstruction
Rene' Girard
Stanford French and Italian
<et.mmt@forsythe.stanford.edu>
Professor Girard will discuss the relation between his views on culture and
its origins and deconstruction. Faculty, graduate students and undergraduates
are all encouraged to attend.
____________
STASS SEMINAR
on Thursday, 4 May
10:00 a.m., Cordura Hall, Room 100
Interpretive Grammar and Stylized Documents
Keith Devlin
CSLI and St. Mary's College
<devlin@csli.stanford.edu>
This is an extension of the talk I gave jointly with Duska Rosenberg in
January. I will outline a grammar that applies to stylized documents
consisting of both constrained syntax slot-and-filler fields and natural
language fields, which interact in complex ways. The grammar associates to
the document situation-theoretic structures that correspond to the strategies
a reader must use in order to interpret the document. This is work in
progress, not a polished performance of a completed project, and I am looking
for suggestions as to how to proceed next. The talk will be self-contained,
but a basic knowledge of situation theory will be assumed (situations, types,
and constraints, and how these are used in the study of information flow).
____________
CSLI SEMINAR
SERIES ON INTELLIGENT AGENTS
on Thursday, 4 May
12:15 p.m., Cordura Hall, Room 100
Perception, Action, and Cognition in an
Integrated Architecture for Physical Agents
Pat Langley
Stanford Robotics Laboratory
<langley@cs.stanford.edu>
In this talk I will describe Icarus, an architecture that integrates sensing,
execution, and planning in sensori-motor domains. The framework represents
control knowledge in terms of durative states and plans that specify sequences
of states. Icarus organizes both states and plans in concept hierarchies,
which it uses to sort experience during the recognition process. The
execution of states can occur in the absence of planning, though the latter
process can modulate execution. Learning involves the alteration of
probabilities about sensor conditions, successor states, and state durations,
which can in turn reduce the amount of sensing during execution. I will
consider Icarus' operation in the domains of pole balancing and flight
control, then discuss its psychological status, its relation to other
integrated architectures, and some directions for future work.
NOTE: A complete schedule for the CSLI seminar series on intelligent agents is
available at URL <http://www-csli.stanford.edu/>. These talks are scheduled
for 12:15 (usually) on Thursdays throughout the Spring quarter. If you order
beforehand, we will provide a sandwich and beverage (for $4.50, paid at the
door; quarters and small bills preferred). Send sandwich orders (turkey, ham,
roastbeef, or vegetarian) to <bocata@csli.stanford.edu> some time on or before
Tuesday of the week of the given talk.
____________
SYMBOLIC SYSTEMS FORUM
on Thursday, 4 May
4:15 p.m., Building 60, Room 61-F
The Virtues of Lightness
Johan van Benthem
Amsterdam and Stanford Philosophy
<johan@csli.stanford.edu>
In his well-known lectures on keywords for the next millennium, Italo Calvino
included "lightness" and "exactness". I want to discuss the combination of
these two virtues, and raise the question how exact logical tools can be
"deconstructed" to provide varying degrees of lightness in modeling reasoning.
By fine-tuning our formal apparatus, we can see more of the essential
complexity of the natural phenomena being modeled. I will demonstrate this
style of analysis in the semantics of standard predicate logic, the general
purpose tool of modern logic. We find a whole landscape of different logics
underneath -- many of them decidable -- reflecting negotiable properties of
the process of "value change" for variables.
____________
LOGIC LUNCH
on Friday, 5 May
12:00 noon, Building 380, Room 383-N
Parametricity and Variants of Girard's J Operator
John Mitchell
Stanford Computer Science
<jcm@Iswim.stanford.edu>
The impredicative polymorphic lambda calculus, or System F, is generally
recognized as a calculus of parametric polymorphism. Intuitively, this means
that the polymorphic functions definable in the language must use the same
algorithm at all types. There are no non-parametric functions such as a
single multiplication function that computes an inner product on vectors,
ordinary product on natural numbers and similar or unrelated ``products'' on
other types.
In his original paper, Girard discusses parametricity in System F and gives an
example showing that if we add a non-parametric operator to the calculus,
normalization fails. This example has been interpreted as a demonstration
that System F is inherently a calculus of parametric functions. Specifically,
since adding a non-parametric operator alters a fundamental property of the
system, we can conclude the System F is inherently parametric. However, this
interpretation of the example relies on the implicit assumption that J is as
simple a non-parametric operator as possible. If it is possible to add a
simpler non-parametric operator to System F in a manner that preserves strong
normalization, we could not really say that Girard's example convincingly
demonstrates the parametric nature of System F.
In this informal talk, we will look at some variants of Girard's example,
hoping to more clearly identify the kind of non-parametric operations that
invalidate normalization. In addition, we observe that impredicativity
appears essential. The talk is based a short note coauthored with Robert
Harper. Since the technical arguments are essentially routine, the main focus
of the talk will be on the intuitive idea of parametricity and its
consequences.
____________
SEMINAR ON HUMAN-COMPUTER INTERACTION
on Friday, 5 May
12:30 p.m., Skilling Auditorium
The Future of the Internet
Marc Andreesen
Netscape Communications Corporation
<marca@netscape.com>
This talk looks at the original concepts behind graphical browsers for the
Internet, which the speaker developed as a student at University of Illinois.
It proceeds with a desciption of what Netscape is addressing in its current
products in order to make the Internet a medium for secure information
exchange and commerce.
MARC ANDREESSEN, 23, is Vice President of Technology and co-founder of
Netscape Communications Corporation in Mountain View, California. Marc
founded the company in April 1994 with Dr. James Clark, founder of Silicon
Graphics, Inc. As an undergraduate at the University of Illinois, Marc
created the NCSA Mosaic' research prototype for the Internet with a team of
students and staff at the university's National Center for Supercomputing
Applications (NCSA). NCSA Mosaic in just over one year gained an estimated
two million users worldwide.
In his role at Netscape Communications, Marc sets and oversees the technical
direction of the company. In December 1994, Netscape Communications released
its first open software products, Netsite server software -- including the
Netsite Commerce Server, the first secure server for the Web -- and Netscape
Navigator client software. Today, Netscape Navigator accounts for more than
65 percent of browser usage on the Web, according to statistics from popular
World Wide Web sites.
Marc was named in 1994 as one of the top 50 people under the age of 40 by Time
Magazine, and was named "Man of the Year" by MicroTimes Magazine. Netscape
Communications was selected as one of the "Hot Companies in 1995" by
Information Week Magazine.
____________
PHILOSOPHY COLLOQUIUM
on Friday, 5 May
3:15 p.m., Building 90, Room 91-A
Physicalism and Physics
Yair Guttmann
Stanford Philosophy
<guttmann@csli.stanford.edu>
____________
LINGUISTICS COLLOQUIUM
on Friday, 5 May
3:30 p.m., Cordura Hall, Room 100
Grammatical Weight
Thomas Wasow
Stanford Linguistics
<wasow@csli.stanford.edu>
Long and complex -- or "heavy" -- elements tend to be placed near the ends of
sentences (as has been known since at least the early years of this century).
I will address three questions concerning this tendency:
(A) What structural and/or pragmatic factors contribute
to it? That is, how should grammatical weight be defined?
(B) Why are heavy elements saved for the ends of sentences?
(C) How should the tendency be captured in a generative
grammar?
Many answers to (A) have been proposed in the literature, but almost none have
been tested systematically. Drawing on numerous examples of heavy NP shift,
particle movement, and the dative alternation from on-line corpora, I will
compare several possible measures of weight.
Speculations regarding (B) have centered around the idea that saving heavy
elements for the end might facilitate parsing. I will discuss this idea,
particularly the version recently worked out in detail by Hawkins, and will
argue that considerations of production may play a role in the phenomenon as
well.
Finally, I will briefly address (C) with respect to heavy NP shift, arguing
against both the traditional rightward movement analysis and the recent
leftward movement approach put forward by Larson and Kayne.
____________
INTENSIONAL LOGIC LECTURE
on Monday, 8 May
1:15 p.m., Building 90, Room 92-Q
Modal Phrase Structure Grammar
Johan van Benthem
Amsterdam and Stanford Philosophy
<johan@csli.stanford.edu>
This is the first of a series of talks on recent developments in and around
modal logic by some guest speakers, illustrating connections with linguistics,
philosophy, computer science, and AI. The first part of this quarter has been
devoted to basic techniques: models and bisimulation, frame correspondences,
modal deduction, completeness arguments, filtration and finite model property.
All interested persons are welcome! All sessions will be in 90-92Q (upstairs
colloquium room, Philosophy Department), Mondays & Wednesdays 1.15 - 2.30 hrs.
May 8 Modal Phrase Structure Grammar Johan van Benthem
10 Relevant Logic Phil Kremer
15,17 Two Lectures on Object Theory Ed Zalta
22 Processes and Bisimulation Rob van Glabbeek
24 Modal theories of Context Sasa Buvac
31 Modal Theorem Proving Grisha Mints
June 5 Dynamic Modal Logic Jan Jaspars
____________
INTENSIONAL LOGIC LECTURE
on Wednesday, 10 May
1:15 p.m., Building 90, Room 92-Q
Relevant Logic
Phil Kremer
Stanford Philosophy
<kremer@csli.stanford.edu>
____________
CSLI SEMINAR
SERIES ON INTELLIGENT AGENTS
on Thursday, 11 May
12:15 p.m., Cordura Hall, Room 100
Toward Building an Android: Recent Developments of the Cog Project
Cynthia Ferrell
MIT AI Laboratory
<ferrell@ai.mit.edu>
Building an android, an autonomous robot of humanoid form coupled with human
like cognitive processes, has been a fantasy of science fiction and a long
standing goal of Artificial Intelligence. In his 1948 paper, "Intelligent
Machinery" (not published until 1970), Alan Turing outlined his view of how to
make computers intelligent. He carefully considered the question of
embodiment, but acknowledged that this route to intelligence was infeasible
given the available technology of the time period. Consequently, he concluded
that the best domains to explore the mechanims of thought were those which
required little contact with the outside world, such as various games and
cryptanalysis.
Modern computational, sensing, and mechanical technologies have improved
dramatically since the time of Turing, and the possibility of building an
android is becoming more and more plausible. The Cog Project is an attempt to
address this long standing challenge. It has two goals: (1) an engineering
goal of building a robot, Cog, that resembles a human in form and function,
and (2) a scientific goal of understanding human cognition. Given the nature
of this challenge, our research draws from diverse scientific influences such
as Artificial Intelligence, Cognitive Science, Ethology, evolutionary
considerations, Neuropsychology, Society of Mind, Connectionism, and Robotics.
The Cog Project is in its infancy, and its members are pursuing a wide variety
of research topics. This talk presents an smorgasbord of current work. Topics
to be covered include Cog's system design (mechanical, computational,
perceptual), and early results in vision, audition, and learning.
CYNTHIA FERRELL is a graduate student at the MIT Artificial Intelligence
Laboratory. She has worked for Professor Rodney Brooks for several years,
receiving her MS degree in May 1993. Her Masters' work focused on autonomous
control and fault tolerant behavior of a small insect-like robot. Currently
she is pursuing her PhD, and is the senior graduate student leading the Cog
Project.
NOTE: A complete schedule for the CSLI seminar series on intelligent agents is
available at URL <http://www-csli.stanford.edu/>. These talks are scheduled
for 12:15 (usually) on Thursdays throughout the Spring quarter. If you order
beforehand, we will provide a sandwich and beverage (for $4.50, paid at the
door; quarters and small bills preferred). Send sandwich orders (turkey, ham,
roastbeef, or vegetarian) to <bocata@csli.stanford.edu> some time on or before
Tuesday of the week of the given talk.
____________
SYMBOLIC SYSTEMS FORUM
on Thursday, 11 May
4:15 p.m., Building 60, Room 61-F
Recent Developments of the Cog Project
Cynthia Ferrell
MIT AI Laboratory
<ferrell@ai.mit.edu>
____________
SEMINAR ON HUMAN-COMPUTER INTERACTION
on Friday, 12 May
12:30 p.m., Skilling Auditorium
The Juno-2 Constraint-Based Drawing Editor
Allan Heydon
DEC Research
<heydon@pa.dec.com>
Constraints are an important enabling technology for interactive graphics
applications. However, today's constraint-based systems are plagued by
several limitations, and constraints have yet to live up to their potential.
Juno-2 is a constraint-based double-view drawing editor that addresses some of
these limitations. Constraints allow you to specify locations in your drawing
declaratively. The constraints are maintained whenever part of the picture is
changed, so constraints make it easier to maintain a picture in the face of
modifications. Some constraints are pre-defined by the Juno-2 application,
but the program also includes a powerful extension language that allows users
to define new constraints. The system demonstrates that fast constraint
solving is possible with a highly extensible, fully declarative constraint
language.
The talk will include a videotape demonstration of Juno-2, including some of
our recent experiments using it to model constrained three-dimensional shapes,
and to produce animations.
This is joint work with Greg Nelson.
ALLAN HEYDON works at Digital Equipment Corporation's Systems Research Center
in Palo Alto, CA. He is a co-developer of Juno-2, a constraint-based drawing
editor. More recently, he has been working on the Vesta software configuration
management system. His general research interests include formal methods,
visual specification languages, programming development environments, and user
interfaces. He received his PhD in computer science from Carnegie Mellon
University in 1991, where he designed and implemented a system for processing
visual specifications of file system security.
____________
PHILOSOPHY COLLOQUIUM
on Friday, 12 May
3:15 p.m., Building 90, Room 91-A
Title to be announced
Marcia Cavell
____________
LINGUISTICS COLLOQUIUM
on Friday, 12 May
3:30 p.m., Cordura Hall, Room 100
Title to be announced
Jaye Padgett
UC Santa Cruz
<padgett@ling.ucsd.edu>
_____________