[Date Prev][Date Next][Date Index]
CSLI Calendar, 8 December 1999, vol. 15:12
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
______________________________________________________________________
8 December 1999 Stanford Vol. 15, No.12
______________________________________________________________________
A weekly publication of the
Center for the Study of Language and Information (CSLI)
Stanford University, Ventura Hall, Stanford, CA 94305-4115
____________
ACTIVITIES FROM 8 DECEMBER TO 31 DECEMBER 1999
WEDNESDAY, 8 DECEMBER
4:00pm Geometric Analysis Seminar
Building 380:381T
Linear and Nonlinear Aspects of Ginzburg-Landau
Vortices
Frank Pacard
Universite Paris 12
http://math.stanford.edu/~moore/ga-sem.html
Abstract below
4:15pm Statistics Seminar
Sequoia Hall 200
An Exponential Family Random Scheme for
Non-Parametric Priors
Sonia Petrone
Universita dell'Insubria - Varese, Italy
http://www-stat.stanford.edu/seminars/seminars.html
Abstract below
THURSDAY, 9 DECEMBER
3:15pm Colloquium
Building 380:380D
The Geometry of Spaces of Phylogenetic Trees
Karen Vogtmann
Cornell University
http://math.stanford.edu/html/seminars.html
4:00pm IMTV Talk
Gates 104
Perceptual Intelligence
Statistical Modeling of Human Behavior
Nuria Oliver
M.I.T.
http://www-forum.stanford.edu
Abstract below
4:00pm Xerox PARC Forum
George Pake Auditorium at Xerox PARC
An Analog Peasant Confronts the Computer Age
Jim Williams
Linear Technology Corp.
http://www.parc.xerox.com/ops/projects/forum/
Abstract below
4:15pm Seminar on Computational Learning and Adaptation
(SCLA)
Cordura 100
Using Correspondence Analysis to Combine Classifiers
Christopher J. Merz
Computational Science Division
Nasa Ames Research Center
http://www-csli.stanford.edu/cll/fall99/merz.html
Abstract below
4:15pm Stanford Algorithms Seminar
Gates 498
Sorting by Reversals is Hard to Approximate
Within Certain Constant
Marek Karpinski
University of Bonn
http://Theory.Stanford.EDU/~aflb/
Abstract below
FRIDAY, 10 DECEMBER
3:15pm Philosophy Colloquium
Building 90:92Q
Skepticism and Naturalism in Hume's Epistemology
Graciela de Pierris
Philosophy, Indiana University
http://www-philosophy.stanford.edu/ce.html#coll
2:30pm Informal Geometry and Topology Seminar
Building 380:383NN
Contact Homology of Coverings II
Klaus Mohnke
http://math.stanford.edu/html/seminars.html
3:30pm Semantics Workshop
Margaret Jacks Hall 126
A Bipartite View of Verb Meaning
Beth Levin (Stanford University)
To be announced
Charles Fillmore (U.C. Berkeley)
http://campus-calendar.stanford.edu/semantics/
Abstract below
SATURDAY, 11 DECEMBER
all day Computer Supported Collaborative Learning
(CSCL'99)
Pre-Conference Workshops
For more information, browse to:
http://learninglab.stanford.edu/CSCL99/
SUNDAY, 12 DECEMBER
1:00pm Computer Supported Collaborative Learning
(CSCL'99)
Start of Conference
For more information, browse to:
http://learninglab.stanford.edu/CSCL99/
MONDAY, 13 DECEMBER
all day Computer Supported Collaborative Learning
(CSCL'99)
For more information, browse to:
http://learninglab.stanford.edu/CSCL99/
12:00pm Brain Research Seminar
Munzer Auditorium, Beckman B060
Timing and Cell Number Control on Neural Development
Martin Raff
University College, London
http://cmgm.stanford.edu/cgi-bin/finger?sem30
2:00pm International Computer Science Institute Talks
ISCI 607 (Berkeley)
Video Multicast using Layered FEC and
Scalable Compression
Wai-tian (Dan) Tan
UC Berkeley
http://www.icsi.berkeley.edu/talks/Wai-tian.html
Abstract below
TUESDAY, 14 DECEMBER
all day Computer Supported Collaborative Learning
(CSCL'99)
For more information, browse to:
http://learninglab.stanford.edu/CSCL99/
4:45pm iBME Curriculum Planning Committee
Teaching Center in the Science & Engineering
Quad
iMBE Curriculum Development Meeting
http://calendus.stanford.edu/CS/read/month.pl
Abstract below
WEDNESDAY, 15 DECEMBER
all day Computer Supported Collaborative Learning
(CSCL'99)
For more information, browse to:
http://learninglab.stanford.edu/CSCL99/
____________
ANNOUNCEMENT
Please note that this is the last calendar of 1999.
____________
GEOMETRIC ANALYSIS SEMINAR
on Wednesday, 8 December 1999, 4:00pm to 5:00pm
Building 380, 381T
http://math.stanford.edu/~moore/ga-sem.html
Linear and nonlinear aspects of Ginzburg-Landau
Vortices
Frank Pacard
We present a joint work with Tristan Riviere concerning existence and
uniqueness questions for Ginzburg-Landau vortices. More precisely, we
describe precisely some branches of critical points of the
Ginzburg-Landau functional \[ E(u) = \int |\nabla u|^2 + \frac{1}{2
\e^2} \int (1 - |u|^2)^2, \] as the parameter $\e$ tends to $0$, here
$u$ is a complex valued function defined in some bounded domain of
${\mathbb R}^2$. In particular we prove that, provided $\e$ is small
enough, all solutions of \[ \Delta u + \frac{u}{\e^2} (1- |u|^2) =0,
\] which are defined in the unit ball and have boundary data given by
$u = e^{i \theta}$ are " radialy symmetric", which means that they are
of the form $u = S (r) \, e^{i \theta}$. Applications to the gauge
invariant Ginzburg-Landau functional are also given.
Dinner: We will be taking the speaker out to dinner
at a nearby restaurant after the seminar.
____________
STATISTICS SEMINAR
on Wednesday, 8 December 1999, 4:15pm
Sequoia Hall 200
http://www-stat.stanford.edu/seminars/seminars.html
An Exponential Family Random Scheme for Non-Parametric Priors
Sonia Petrone
Universita' dell'Insubria - Varese. Italy
For Bayesian semi- and non-parametric inference, we need a prior on a
``large'' class of distribution functions on the sample space; and in
applications where we think of the data as continuous, we want a
nonparametric prior which selects an absolutely continuous
distribution function.
In this work we describe one way of constructing a nonparametric
prior with the required properties, based on the notion of
``Feller-type approximation.'' Roughly speaking, we model the density
of the data as a mixture of given densities. The components of the
mixture have no unknown parameters and are related to the natural
exponential family. The mixing weights and the number of components of
the mixture (or the order of approximation) are unknown and have a
prior distribution.
Applications include density estimation, estimating a mixing
distribution and non parametric regression. A nice property of the
proposed estimators is that the choice of the smoothing parameter is
driven by the data through its posterior distributions. This is still
work in progress and examples and results about consistency of the
posterior will be restricted to the case of data in [0,1].
Joint work with Piero Veronese and Larry Wasserman.
____________
IMTV TALK
on Thursday, 9 December 1999, 4:00pm
Gates 104
http://www-forum.stanford.edu
Perceptual Intelligence:
Statistical Modeling of Human Behavior
Nuria Oliver
http://nuria.www.media.mit.edu/people/nuria/
In the talk I will present my work towards building Perceptually
Intelligent systems. During the last years at the MIT Media Lab I have
been developing a framework for the automatic recognition of different
kinds of human behavior from video cameras and other sensors. In
particular, I propose a statistical machine learning framework for
capturing interactions between several agents --humans or cars--. In
the case of humans two Hidden Markov Models (HMMs) are coupled (CHMMs)
to capture the interactions between them. In the case of cars, a
lattice of 6 CHMMs is proposed for modeling the pairwise mutual
interactions between adjacent cars.
Three systems that use the proposed paradigm are presented: (1)
LAFTER, an automatic face detection and tracking system with facial
expression recognition; (2) a visual surveillance system that
recognizes human to human typical interactions; (3) and a Smart Car
that recognizes driver behavior.
These models would let us categorize human actions very soon after the
beginning of the action. Because of the generic nature of the typical
behaviors of each of the implemented systems there is a reason to
believe that this approach to modeling human behavior would generalize
to other dynamic human-machine systems. This would allow us to
recognize automatically people's intended action, and thus build
control systems that dynamically adapt to better suit human's
purposes.
____________
XEROX PARC FORUM
on Thursday, 9 December 1999, 4:00pm to 5:00pm
George Pake Auditorium at Xerox PARC
http://www.parc.xerox.com/ops/projects/forum/
An Analog Peasant Confronts the
Computer Age
Jim Williams
Linear Technology Corp.
This talk considers the place of the analog circuit designer in a
world (seemingly) committed to digital computer technology. Analog
techniques are shown to be very much alive and employed, and unlikely
to obsolesce. The discussion concludes with visually augmented
commentary on CAD and the design process.
Biography: Jim Williams was at the Massachusetts Institute of
Technology from 1968 to 1979, concentrating exclusively on analog
circuit design. His teaching and research interests involved
application of analog circuit techniques to biochemical and biomedical
problems. Concurrently, he consulted U.S. and foreign concerns and
governments, specializing in analog circuits. In 1979, he moved to
National Semiconductor Corporation, continuing his work in the analog
area with the Linear Integrated Circuits Group. In 1982, he joined
Linear Technology Corporation as staff Scientist, where he is
presently employed. Interests include product definition, development,
and support. Jim has authored over 250 publications relating to analog
design. He received the 1992 Innovator of the Year Award from EDN
magazine for work in high-speed circuits. His spare time interests
include sports cars, collecting antique scientific instruments, art,
and restoring and using old Tektronix oscilloscopes. He lives in Palo
Alto, California with his wife, son, a dog named Bonillas, and 28
Tektronix oscilloscopes.
____________
SEMINAR ON COMPUTATIONAL LEARNING AND ADAPTATION
on Thursday, 9 December 1999, 4:15pm to 5:30pm
Cordura 100
http://www-csli.stanford.edu/cll/fall99/merz.html
Using Correspondence Analysis to Combine
Classifiers
Christopher J. Merz
Computational Science Division
NASA Ames Research Center
Several effective methods have been developed recently for improving
predictive performance by generating and combining multiple learned
models. The general approach is to create a set of learned models
either by applying an algorithm repeatedly to different versions of
the training data, or by applying different learning algorithms to the
same data. The predictions of the models are then combined according
to a voting scheme. This talk focuses on the task of combining the
predictions of a set of learned models. The method described uses the
strategies of stacking and Correspondence Analysis to model the
relationship between the learning examples and their classification by
a collection of learned models. A nearest neighbor method is then
applied within the resulting representation to classify previously
unseen examples. The new algorithm does not perform worse than, and
frequently performs significantly better than other combining
techniques on a suite of data sets.
____________
STANFORD ALGORITHMS SEMINAR
on Thursday, 10 December 1999, 4:15pm
Gates Building 498
http://Theory.Stanford.EDU/~aflb/
Sorting by Reversals is Hard to Approximate
Within Certain Constant
Marek Karpinski
University of Bonn
We prove that the problem MIN-SBR of sorting a permutation by the
minimum number of reversals is hard to approximate (NP-hard by
randomized reductions) within any constant factor less than some
explicit threshold. This excludes an existence of a PTAS for this
problem under usual assumptions, thus settling a question which was
open for some time. The proof method uses certain new explicit
approximation hardness techniques for bounded dependency, and bounded
degree optimization problems. The MIN-SBR problem has been motivated
and extensively studied in computational molecular biology, but
existence of PTASs remained an open issue. This problem connects also
to the well known problem of sorting by prefix reversals (or, so
called, pancake sorting). Our nonapproximability result for MIN-SBR is
also in sharp contrast to its signed version for which efficient exact
algorithms have been designed recently.
(Joint work with P.Berman.)
____________
SEMANTICS WORKSHOP
on Friday, 10 December 1999, 3:30pm
Margaret Jacks Hall 126
http://campus-calendar.stanford.edu/semantics/
A Bipartite View of Verb Meaning
Beth Levin
Stanford University
Many researchers in lexical semantics either implicitly or explicitly
make a distinction between two aspects of verb meaning, which I term
the "structural" and the "idiosyncratic" (Grimshaw 1993; Hale and
Keyser 1993; Jackendoff 1990, 1996; Rappaport Hovav and Levin 1995,
1998; Pinker 1989; among others).
The structural aspects of verb meaning are those that define various
ontological types of events and, consequently, are often referred to
as event structures. They are also said to be the
grammatically-relevant aspects; they define the semantic classes of
verbs whose members share syntactically and morphologically-salient
properties. In contrast, the idiosyncratic facets of verb meaning,
what might also be called "core" meaning, serve to differentiate a
verb from other verbs sharing the same structural aspects of meaning.
In this talk, I will examine evidence for distinguishing these two
facets of verb meaning. I will then discuss the relationship between
them, as it has barely been investigated in previous studies, and I
will propose that the idiosyncratic components constrain the
structural components. I will also argue that the idiosyncratic
component has a part to play in argument expression. Although most
arguments are licensed by the structural components of verb meaning,
some are licensed only by the idiosyncratic components of verb
meaning.
Time permitting, I will discuss how the recognition of this form of
argument licensing can shed light into some of the challenges of
"objecthood": the significant variation both within and across
languages as to the set of NPs identified as objects and the lack of
uniform semantic characterization of all objects, despite the
agreement that prototypical objects are "patients".
____________
INTERNATIONAL COMPUTER SCIENCE INSTITUTE TALKS
on Monday, 13 December 1999, 2:00pm to 3:30pm
ISCI 607 (Berkeley)
http://www.icsi.berkeley.edu/talks/Wai-tian.html
Video Multicast using Layered FEC and
Scalable Compression
Wai-tian (Dan) Tan
UC Berkeley
The use of scalable video with layered multicast has been shown to be
an effective method to achieve rate control in heterogeneous networks.
In this talk, we discuss the use of layered FEC as an error control
mechanism that allows receivers to individually trade-off latency for
received video quality.
While FEC-based error control methods are relieved from the ACK/NACK
implosion problems of ARQ-based schemes, the use of layered FEC with
scalable compression has several other advantages. First, it permits
rate control and allows FEC to be used without overall data
expansion. Second, unequal error protection can be conveniently
employed. Third, FEC packets are multicast only to receivers that need
them, thereby conserving network bandwidth.
The talk will cover an overview of the basic layered FEC scheme and
discuss some practical issues such as choosing "good" scalable
compression methods, rate control algorithms and error control codes
for video multicast.
This talk will be held in the Main Lecture Hall at ICSI. 1947 Center
Street, Sixth Floor, Berkeley, CA 94704-1198 (on Center between Milvia
and Martin Luther King Jr. Way)
____________
iBME PLANNING COMMITTEE
on Tuesday, 14 December 1999, 4:45pm to 6:30pm
Teaching Center in the Science & Engineering Quad
http://calendus.stanford.edu/CS/read/month.pl
iBME Curriculum Development Meeting
Open to Stanford Students & Faculty - not an MDN Event
On behalf of the Institute for Biomedical Engineering (iBME) Planning
Committee, you are invited to participate in the first meeting for the
Stanford faculty and students to discuss planning for iBME curriculum
development and strategy.
Please join us on Tuesday, December 14th in the Teaching Center in the
Science and Engineering Quad (TCSEQ) at 5 pm. See below to view the
preliminary agenda and directions to the TCSEQ.
Preliminary Agenda
iBME Curriculum Development Planning Meeting
Tuesday, December 14, 1999
TCSEQ 201, Bloch Lecture Hall
4:40 Reception - TCSEQ Foyer
5:00 Introduction
Thomas Andriacchi, PhD
Paul Yock, MD
5:05 Review of Other U.S. Academic BME Programs
5:30 The Whitaker Opportunity
5:40 Existing BME Courses at Stanford
6:00 Proposed Structure for New Curriculum
6:15 Process for Curriculum Development
6:30 Adjourn
The TCSEQ is located on Serra Mall and can be viewed on the Stanford
online campus map by going to:
http://www.stanford.edu/dept/registrar/tcseq/map.html
For those not able to attend the session on the 14th, please note that
a summary of the meeting will be posted to the iBME website by
December 21st. Go to http://ibme.stanford.edu to view the iBME website
(accessible to Stanford personnel only).
____________
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
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.
and
news://nntp.stanford.edu/su.events
Information about CSLI's research program is available at
http://www-csli.stanford.edu/.
For maps to the Stanford University campus see
http://www.stanford.edu/home/visitors/maps.html.
____________