[Prev][Next][Index]

CSLI Calendar, 07 Dec 1995, vol.11:10




        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
______________________________________________________________________________

7 December 1995                   Stanford                     Vol. 11, No. 10
______________________________________________________________________________

      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 6 -- 15 DECEMBER 1995

  WEDNESDAY, 6 DECEMBER
         3:15 - Philosophy of Computation Seminar
                Ventura Hall, Room 17
                Brian Cantwell Smith, Xerox PARC
                Abstract below

  THURSDAY, 7 DECEMBER
        10:00 - STASS Seminar
                Cordura Hall, Room 100
                Barwise-Seligman Discussion Group
                Keith Devlin, CSLI / St.Mary's College
                Abstract below

         4:15 - SSP Forum
                Building 60, Room 61-F
                Better than the Best: The Power of Cooperation
                Bernardo A. Huberman,  Xerox PARC
                Abstract below

         4:15 - Seminar on Computational Learning and Adaptation
                Cordura Hall, Room 100
                Using Evidence Grids for Place Learning in Dynamic Environments
                Brian Yamauchi, ISLE and Stanford University
                Abstract below

  FRIDAY, 8 DECEMBER
        12:00 - Logic Lunch
                Building 380, Room 383-N
                Admissible Inference Rules for Nonstandard Logics
                Vladimir V. Rybakov
                Abstract below

        12:30 - HCI Seminar
                Skilling Auditorium, SITN Channel E1
                Design for Universal Access
                Betsy Bayha, World Institute on Disability
                Abstract below

  WEDNESDAY, 13 DECEMBER
	8:00pm- 1995-96 Immanuel Kant Lectures, 1
		Medical School, Fairchild Auditorium
		The Last Word: Two Lectures on Reason
		Thomas Nagel, New York University

  THURSDAY, 14 DECEMBER
	8:00pm- 1995-96 Immanuel Kant Lectures, 2
		Medical School, Fairchild Auditorium
		The Last Word: Two Lectures on Reason
		Thomas Nagel, New York University

  FRIDAY, 15 DECEMBER
	 3:15 - 1995-96 Immanuel Kant Lectures
		Encina Hall, Room 423
		Discussion Seminar
		Thomas Nagel, New York University

                               ____________

The CSLI Calendar appears on Wednesday of each week throughout the academic
year.  Announcements, abstracts, and other information to appear in the
Calendar can be submitted by e-mail to [incalendar@csli.stanford.edu].

Further information about CSLI and past issues of the CSLI Calendar
are available on the Internet at URL [http://www-csli.stanford.edu/csli/].
The Calendar is also posted each week to the [csli.bboard] newsgroup.

                               ____________

                    PHILOSOPHY OF COMPUTATION SEMINAR
                        on Wednesday, 6 December
                     3:15 p.m., Ventura Hall, Room 17
                           Brian Cantwell Smith
                    Xerox PARC and Stanford Philosophy
                         [bcsmith@parc.xerox.com]

We will have an _extra-long session_ this week, devoted to presentations by
seminar participants.

                               ____________

                              STASS SEMINAR
                         on Thursday, 7 December
                    10:00 a.m., Cordura Hall, Room 100
                    Barwise-Seligman Discussion Group
                               Keith Devlin
                         CSLI / St.Mary's College
                        [devlin@csli.stanford.edu]

We will continue working through the manuscript of the forthcoming Barwise and
Seligman book on the mathematics of information, concentrating today on
Chapter 2.  Copies of relevant portions of the text may be obtained from Eric
Hammer [ehammer@csli.stanford.edu] at CSLI.  This will be the last meeting of
the STASS Seminar this quarter.  The Barwise-Seligman read-through will
continue next quarter.

                               ____________

                              CSLI COGLUNCH
             on Thursdays, 12:00 noon, Cordura Hall, Room 100
                    [contact: guven@csli.stanford.edu]

There is no CogLunch on December 7, 1995.  We will resume our series during
the second week of classes in the Winter Quarter, on Thursday January 18,
1996.  The theme will continue to be consciousness.  Our first speaker will be
David Spiegel, from the Stanford Psychiatry Department, who will talk on
"Disintegrated Experience: Dissociation, Hypnosis, and Trauma."  Hope to see
you all there!

                               ____________

                          SYMBOLIC SYSTEMS FORUM
                         on Thursday, 7 December
                    4:15 p.m., Building 60, Room 61-F
              Better than the Best: The Power of Cooperation
                           Bernardo A. Huberman
                                Xerox PARC
                        [huberman@parc.xerox.com]

It is widely thought that a group of cooperating individuals can solve
problems faster and better than either a single person or the same group of
individuals working in isolation from each other. Such belief underlies the
founding of the firm, the existence of communities of practice, and the
establishing of groups charged with solving hard problems.

I will present a theory of the performance enhancement that results from
cooperation among agents, and its dynamical consequences for the group as a
whole. The theory predicts a universal law of performance and a specific
evolution path for the community of agents as its members learn and
specialize.  A number of experimental results support these findings, as well
as providing new insights into the dynamics of cooperation.

BERNARDO A. HUBERMAN is a Research Fellow at the Xerox Palo Alto Research
Center, where he heads a group involved in studying the dynamics of
distributed processes in social organizations and computational systems.  He
received his Ph.D. in Physics from the University of Pennsylvania, and is
concurrently a Consulting Professor of Physics and a member of the faculty of
the Symbolic Systems Program at Stanford University.  He has worked in
condensed matter physics and the theory of critical phenomena, and is one of
the discoverers of chaos in a number of physical systems.  He also established
the existence of a number of universal properties in nonlinear dynamics, and
as part of his research on the evolution of complex systems, he discovered the
phenomenon of ultradiffusion in hierarchical structures. In the field of
computation he predicted the existence of phase transitions in artificial
intelligence, which have been observed in a number of computationally hard
problems. He also started the field of ecology of computation, and is the
editor of a book on the subject.

Recently, Dr. Huberman has been studying the power of cooperation in
collective problem solving by many agents, and the dynamics of collective
action and learning in multiagent organizations.  He coauthored an article on
the Dynamics of Social Dilemmas in the March 1994 issue of Scientific
American.

                               ____________

             SEMINAR ON COMPUTATIONAL LEARNING AND ADAPTATION
                         on Thursday, 7 December
                    4:15 p.m., Cordura Hall, Room 100
                 Using Evidence Grids for Place Learning
                         in Dynamic Environments
                              Brian Yamauchi
                       ISLE and Stanford University
                     [yamauchi@flamingo.stanford.edu]

A central problem in mobile robot navigation is localization: How can a robot
determine its current location based upon its sensor data?  In many real-world
environments, this task is complicated by the presence of human beings.  Human
environments are dynamic environments, and in dynamic environments the sensor
data obtained from the same location may change over time.  In this talk, I
will describe a method of place learning based on building local evidence
grids and associating each grid with a place in the world.  Evidence grids
represent space as a Cartesian grid, where each cell in the grid has a certain
probability of being occupied. Place recognition consists of building an
evidence grid for an unknown location and matching this grid against
previously learned grids. A multi-resolution hill-climbing algorithm is used
to search the space of possible translations and rotations between the new
grid and the learned grids.  I will also describe preliminary results obtained
for this place learning algorithm using a real mobile robot in an office
environment containing multiple forms of dynamic change.

This talk describes work done in collaboration with Pat Langley at the
Stanford Robotics Laboratory.

                              _____________

                               LOGIC LUNCH
                          on Friday, 8 December
                   12:00 noon, Building 380, Room 383-N
            Admissible Inference Rules for Nonstandard Logics
                           Vladimir V. Rybakov
                       [rybakov@csli.stanford.edu]

The study of nonstandard logical systems in many respects is founded on the
use of different inference rules.  Usually the study of any logical system
begins with collecting some necessary set of admissible inference rules.
These rules were introduced into consideration by P. Lorenzen (1955) who first
fixed clearly their definition.  These rules form the greatest class of rules
which can be employed in derivations while preserving the class of provable
theorems.  There have been and are existing now a lot of open important
problems concerning applicability of inference rules, and in particular, their
admissibility.

The problematics of admissible inference rules has a direct connection with
questions of structure and properties of quasi-varieties generated by some
algebras from the algebraic semantics for nonstandard logics.

For instance, Harvey Friedman's problem of whether there exists an algorithm
for recognizing admissible inference rules for Intuitionistic Propositional
Calculus is equivalent to the problem of whether the quasi-variety generated
by the free pseudo-boolean algebra of countable rank has a decidable
quasi-equational theory.  And Kuznetsov's problem of whether intuitionistic
logic has a finite base for admissible rules has as an algebraic analog the
problem of whether the free pseudo-boolean algebra of countable rank has a
finite base for quasi-identities.  The Novikov-Mints problem of recognizing
solvable logical equations also has an algebraic analog: the question whether
exists an algorithm for recognizing of solvable equations in appropriate free
algebras.  It is also of interest that there is an approach to this problem by
means of admissibility of inference rules with metavariables.

This talk will contain an overview of some modern results connected with
properties of inference rules, and in particular their admissibility, for
modal and superintuitionistic logics.  The presentation will concentrate on
questions of recognizing admissibility rules, studying bases for admissible
rules, resolving logical equations, structural completeness, etc.

                              _____________

                  SEMINAR ON HUMAN-COMPUTER INTERACTION
                          on Friday, 8 December
                     12:30 p.m., Skilling Auditorium
                             Universal Design
                               Betsy Bayha
                      World Institute on Disability
                             [betsy@wid.org]

Universal design of information systems and telecommunications means
developing systems flexible enough to accommodate the broadest possible range
of users regardless of age or disability.  The time has come to think
creatively, to move beyond the outmoded, clumsy and costly approach of
retrofitting systems with adaptive equipment to create access for many end
users.  Universal design calls for flexible interfaces to be built into the
product at the blueprint stage, producing systems more easily used by
everyone.  Implemented in its best sense, universal design can result in an
expanded market and a mutually beneficial situation for both industry and
consumers.

The presentation will also feature Electronic Curbcuts, a video on universal
design and computer interfaces, which was produced by the Universal Access
Project.

BETSY BAYHA directs the Universal Access Project, an endeavor of the World
Institute on Disability, The Trace Research and Development Center, and the
CPB/WGBH National Center on Accessible Media.  The Universal Access Project,
funded largely by a planning grant from the U.S. Department of Commerce's
National Telecommunications and Information Administration, focuses on
building accessibility for disabled persons into the information superhighway
at the design level.  Through ongoing dialogue with industry, consumer groups,
agencies and related projects, consensus-based guidelines will be developed.

This is the last PCD Seminar talk for the quarter.  We will start again on
Friday, January 12.  Stay tuned for the schedule for next quarter.  Happy
Holidays.

                              _____________

		      1995-96 IMMANUEL KANT LECTURES
		on Wednesday and Thursday, 13--14 December
	       8:00 p.m., Fairchild Auditorium (Med School)
		  The Last Word: Two Lectures on Reason
			  Professor Thomas Nagel
			   New York University

			Also, a discussion seminar
			  on Friday, 15 December
		     3:30 p.m., Encina Hall, Room 423

			      _____________