[Date Prev][Date Next][Date Index]
CSLI Calendar, Wednesday, 12 May 2004, vol. 19:35
CSLI CALENDAR OF PUBLIC EVENTS
______________________________________________________________________
12 May 2004 Stanford Vol. 19, No. 35
______________________________________________________________________
A weekly publication of the
Center for the Study of Language and Information (CSLI)
Stanford University, Cordura Hall, Stanford, CA 94305-4115
http://www-csli.stanford.edu/
____________
ACTIVITIES FROM 12 MAY 2004 TO 21 MAY 2004
WEDNESDAY, 12 MAY 2004
12 noon Psychology Developmental Brownbags
Bldg. 380:381U
"Development of face recognition in school-aged children"
Golijeh Golarai
Psychology
http://www-psych.stanford.edu/news.html#dev_brownbag
3:45pm Psychology Department Colloquium
Jordan Hall 420:041
Title to be announced
Barbara Rogoff
University of California, Santa Cruz
http://psych.ucsc.edu/Faculty/bRogoff.shtml
http://www-psych.stanford.edu/news.html
4:15pm CSLI Seminar on Computational Learning and Adaptation
Cordura Hall, room 100
"A Novel Approach to Novelty Detection"
Jeff Scargle
NASA Ames Research Center
http://cll.stanford.edu/scla.html
Abstract below
4:15pm EE380: Computer Systems Lab Colloquium
Gates B03 (NEC Auditorium)
"Controlling Digital Cloth"
Ari Rapkin
Industrial Light and Magic
http://ee380.stanford.edu/contents.html
Abstract below
(this talk will only be available live)
5:15pm CCRMA Colloquium
CCRMA Ballroom, The Knoll
"Ragtime, Fugue or Madrigal:
Describing Metric Peculiarities with a Computational Approach"
Anja Volk
http://www-ccrma.stanford.edu/
THURSDAY, 13 MAY 2004
12 noon RNI/Stanford Seminar Series on Theoretical Neuroscience
BioX/Clark Center, Room S360
"The Ersatz Brain Project:
Brain-Like Computer Design for Cognitive Applications"
James Anderson
Brown University
http://www.brainscience.brown.edu/departments/faculty/anderson.html
http://www.rni.org/seminar2.html
Abstract below
3:00pm Reuters Foundation Digital Vision Program Seminar
Reuters lounge, Cordura Hall
Rick Smolan
Against All Odds Productions
http://cgi.stanford.edu/group/reuters/cgi-bin/calendar/index.cgi
4:00pm Personality Lab
Jordan Hall 420:419
"The effects of early life stress on socioemotional behavior,
cognition, and the HPA axis"
Karen Parke
Stanford, Psychiatry
http://www-psych.stanford.edu/news.html#person_lab
4:00pm PARC Forum
George Pake Auditorium at PARC
"Actor-Oriented Design: Concurrent Models as Programs"
Edward Lee
University of California, Berkeley
http://www.parc.com/forum/
Abstract below
4:15pm SSP10: Symbolic Systems Forum
Bldg. 380:380C (Math Corner)
"How people coordinate with each other with and without language"
Herbert H. Clark
Psychology, Stanford
http://symsys.stanford.edu:8081/ssp-dynamic/servlet/ssp_events
Abstract below
4:15pm Information Systems Seminar
Packard 202
"Linear Time Universal Coding in the Class of Tree Models via
FSM Closure"
Marcelo Weinberger
HP Labs
http://isl.stanford.edu/groups/seminar/
4:30pm Medicine and the Muse: An Arts, Humanities and Medicine Symposium
Cantor Arts Center Auditorium
Rafael Campo
Author of The Healing Art: A Doctor's Black Bag of Poetry
Harvard Medical School and Beth Israel Deaconess Medical Center
(Exhibit opens at 4:00pm)
http://scbe.stanford.edu/events/muse.html
6:15pm Stanford Phonology Workshop
Margaret Jacks Hall 460:426 (Terrace room)
"The Interaction of Multiple Sources of Information in Korean
Sentence Processing"
Hee-Sun Kim
Stanford University
http://www-linguistics.stanford.edu/pinterest/
Abstract below
FRIDAY, 14 MAY 2004
12:30pm CS547: Human-Computer Interaction Seminar
Gates B01
"Collaboration, Tool Use, and Work Practice of Mars Mission Scientists"
Roxana Wales and Alonso Vera
NASA Ames
http://www-pcd.stanford.edu/cs547/
Abstract below
2:00pm NLP Reading Group
Ventura 17
"Using Curvature and Markov Clustering in Graphs for Lexical
Acquisition and Word Sense Discrimination"
Dominic Widdows
http://www-nlp.stanford.edu/nlp/nlpgroup.html
Abstract below
3:15pm Philosophy Department Colloquium
Bldg. 90:92Q
"Racial Virtues"
Lawrence Blum
University of Massachusetts-Boston
http://www-philosophy.stanford.edu/ce.html
3:15pm Friday Cognitive Seminar
Jordan Hall 420:050
Title to be announced
Nick Davidenko
http://www-psych.stanford.edu/news.html#frisem
SATURDAY, 15 MAY 2004
1:00pm Continuing Studies: The Human Mind
Cubberley Auditorium
"Memory and Learning"
Gordon Bower
Psychology, Stanford
John Gabrieli
Psychology and Neurosciences, Stanford
Anthony Wagner
Psychology, Stanford
http://continuingstudies.stanford.edu/course/EVT66.asp
MONDAY, 17 MAY 2004
all day World Intellectual Property Organization (WIPO) Conference
Stanford Law School
"WIPO Comes to Stanford: High Tech IP Issues in a Global Marketplace"
Two day conference. Pre-registration is required
http://lst.stanford.edu/wipo.
4:15pm CS528: Broad Area Colloquium in AI,
Geometry, Graphics, Robotics, and Vision
TCSeq 200
"Life-Sized Learning"
Leslie Kaelbling
MIT
http://graphics.stanford.edu/courses/cs528/
Abstract below
TUESDAY, 18 MAY 2004
all day World Intellectual Property Organization (WIPO) Conference
Stanford Law School
"WIPO Comes to Stanford: High Tech IP Issues in a Global Marketplace"
Two day conference. Pre-registration is required
http://lst.stanford.edu/wipo/
12 noon State of the School Address
Clark Center Auditorium (BioX)
James D. Plummer
Dean, School of Engineering
4:15pm Logic Seminar
Bldg. 380:380F (math corner)
"Martin-Lof's type theory with permutative reductions"
Makoto Tatsuta
National Institute of Informatics
http://www-logic.stanford.edu/seminars.html
WEDNESDAY, 19 MAY 2004
12 noon Psychology Developmental Brownbags
Bldg. 380:381U
Title to be announced
Hazel Markus, Claude & Dorothy Steele
Psychology, Stanford
http://www-psych.stanford.edu/news.html#dev_brownbag
4:15pm EE380: Computer Systems Lab Colloquium
Gates B03 (NEC Auditorium)
"Nano-Architectures, Nano-Computing, Nano-Technologies and the
DNA Structure"
S. Barbu, M. Morf, A. E. Barbu
ST Microelectronics, Stanford University, UC Davis (Biology)
http://ee380.stanford.edu/contents.html
5:15pm CCRMA Colloquium
CCRMA Ballroom, The Knoll
"Tabla Transcription"
Parag Chordia
CCRMA
http://www-ccrma.stanford.edu/
Abstract below
THURSDAY, 20 MAY 2004
12:15pm CSLI CogLunch
Cordura Hall, Room 100
Title to be announced
Ariane Tom
http://www-csli.stanford.edu/events/Coglunch/
4:00pm Personality Lab
Jordan Hall 420:419
"Creativity in science:
Individual differences and developmental antecedents"
Dean Simonton,
UC Davis
http://www-psych.stanford.edu/news.html#person_lab
4:00pm PARC Forum
George Pake Auditorium at PARC
"Air Traffic Control Systems"
Claire Tomlin
Stanford
http://www.parc.com/forum/
4:15pm SSP10: Symbolic Systems Forum
Bldg. 380:380C (Math Corner)
"Making sense of it all: Pattern prediction in infancy"
Natasha Kirkham
Psychology Department
http://symsys.stanford.edu:8081/ssp-dynamic/servlet/ssp_events
Abstract below
4:15pm Information Systems Seminar
Packard 202
"On Universal Types"
Gadiel Seroussi
HP Labs
http://isl.stanford.edu/groups/seminar/
5:30pm Ethics in Society
Bldg. 260:113
"The Right to Offend?"
A panel on ethics in advertising co-sponsored by The Stanford
Daily and the Ethics in Society Program.
http://www.stanford.edu/dept/EIS/lectures.html
6:15pm Stanford Phonology Workshop
Margaret Jacks Hall 460:126
"The Origin of Algonquian Complex Stem Morphology:
Evidence from Yurok"
Andrew Garrett
http://www-linguistics.stanford.edu/pinterest/
Abstract below
FRIDAY, 21 MAY 2004
all day UC Berkeley Conference on Methods in Phonology
370 Dwinelle (UC Berkeley)
http://linguistics.berkeley.edu/lingdept/Current/events.html
12:30pm CS547: Human-Computer Interaction Seminar
Gates B01
"Giving Users Control in Ubiquitous Computing Environments"
Anind Dey
Intel Research and UC Berkeley.
http://www-pcd.stanford.edu/cs547/
Abstract below
2:00pm NLP Reading Group
Ventura 17
Title to be announced
Kiyoko Uchiyama
http://www-nlp.stanford.edu/nlp/nlpgroup.html
3:00pm Learning, Design & Technology (LDT) Masters presentations
Wallenberg Hall Learning theater (Bldg. 160)
http://www.stanford.edu/dept/SUSE/ldt/projects.htm
Abstract below
3:15pm Friday Cognitive Seminar
Jordan Hall 420:050
Title to be announced
Danny Oppenheimer
http://www-psych.stanford.edu/news.html#frisem
3:30pm Semantics and Pragmatics Workshop
Margaret Jacks Hall 460:126
"Macro-events:
Principles of Event Encoding at the Syntax-Semantics Interface"
Jumlrgen Bohnemeyer
SUNY Buffalo
http://www-linguistics.stanford.edu/semgroup/
Abstract below
7:00pm Stanford Conference
Bldg. 260:113
"Diversity and Universals in Language: The Consequences of Variation"
http://dlcl.stanford.edu/research/workgroups/diversity-conf.html
Information below
SATURDAY, 22 MAY 2004
all day Stanford Conference
Bldg. 300:300T
"Diversity and Universals in Language: The Consequences of Variation"
http://dlcl.stanford.edu/research/workgroups/diversity-conf.html
Information below
all day UC Berkeley Conference on Methods in Phonology
370 Dwinelle (UC Berkeley)
http://linguistics.berkeley.edu/lingdept/Current/events.html
SUNDAY, 23 MAY 2004
all day Stanford Conference
Bldg. 300:300T
"Diversity and Universals in Language: The Consequences of Variation"
http://dlcl.stanford.edu/research/workgroups/diversity-conf.html
Information below
all day UC Berkeley Conference on Methods in Phonology
370 Dwinelle (UC Berkeley)
http://linguistics.berkeley.edu/lingdept/Current/events.html
____________
Stanford Blood Center status: Shortage of O+, O-, A+, and A-. For an
appointment: http://bloodcenter.stanford.edu/ or call 650-723-7831.
It only takes an hour of your time.
____________
ANNOUNCEMENT
Stanford University officially broke ground Tuesday 11 May 2004 for
the new CSLI/EPGY annex (to be located approximately on the old
parking lot). See http://cpm.stanford.edu/CSLI/ for more information.
____________
CSLI SEMINAR ON COMPUTATIONAL LEARNING AND ADAPTATION
on Wednesday, 12 May 2004, 4:15pm to 5:30pm
Cordura 100
http://cll.stanford.edu/scla.html
"A Novel Approach to Novelty Detection"
Jeff Scargle
Space Science Division, NASA Ames Research Division
In this talk, I present a brief overview of the problem of detecting
novel events, outliers, and anomalies, then discuss the effectiveness
of Voronoi tessellations of multiparametric data spaces and their
relevance to this problem. After this, I describe a new, simple
algorithm for identifying outliers that takes advantages of these
structures. I demonstrate results on tasks taken from the literature,
including medical diagnosis and ball-bearing fault detection. But the
application that inspired this work is the prompt detection of
anomalies in water distribution systems, which has applications in
maintaining water quality and recognizing bioterrorism events, so I
discuss its use there as well.
____________
EE380: COMPUTER SYSTEMS LAB COLLOQUIUM
on Wednesday, 12 May 2004, 4:15pm to 5:30pm
NEC Auditorium (B03), Gates Computer Science Building
http://ee380.stanford.edu/contents.html
"Controlling Digital Cloth"
Ari Rapkin
Industrial Light + Magic
Over the last few years, cloth simulation has become a major tool for
creating believable visual effects. Computer generated characters wear
complex clothing, sports arenas are draped with banners and flags, and
non-cloth items like hair, wings, and plastic bags can be simulated
with the same numerical methods. The physics behind these simulations
is improving every year, and it's possible to achieve superb fidelity.
However, when making a film, it's not enough to get the physics right.
Sometimes getting the physics right is the wrong thing entirely! The
artists need to be able to override the physics, making the cloth
behave as needed to convey the story they're trying to tell. A key
feature of ILM's cloth simulation system is the user's ability to
balance physical correctness with artistic and directorial control.
I'll present examples of digital cloth from several recent films, and
provide a closer look at some of the ways ILM's visual effects artists
control simulated cloth to achieve their desired results.
About the speaker: Ari Rapkin began working at Industrial Light +
Magic in 1998 as a member of the Production Software team, providing
support and development for a variety of graphics software systems. In
2000, she joined the Software R&D department's simulation group. Her
contributions to ILM's software include fluid & smoke simulation for
films including Pearl Harbor and The Mummy Returns. Since early 2002,
she has been the head of the cloth simulation project, working on
cloth simulation for films such as Jurassic Park III, Star Wars:
Episodes II and III, Hulk, Pirates of the Caribbean, Harry Potter and
the Chamber of Secrets, Harry Potter and the Prisoner of Azkaban, and
Van Helsing.
Ari was born in Los Angeles, but soon moved to Texas and then to
Wilmington, Delaware. She attended The Johns Hopkins University in
Baltimore, where she received a B.A. in mathematics. Her graduate
education began with a program in gifted education at the University
of Virginia, where she also earned a master's degree in computer
science. Later she went on to obtain an M.S. in computer science at
Carnegie Mellon University, graduating in 1997. While in graduate
school, she was a math instructor at the Johns Hopkins Center for
Talented Youth, and a research intern at Xerox PARC and DEC SRC.
____________
RNI/STANFORD SEMINAR SERIES IN THEORETICAL NEUROSCIENCE
on Thursday, 13 May 2004, 12 noon - 1pm
BioX/Clark Center, Room S360
http://www.rni.org/seminar2.html
"The Ersatz Brain Project:
Brain-Like Computer Design for Cognitive Applications"
James Anderson
Cognitive and Linguistic Sciences, Brown University
http://www.brainscience.brown.edu/departments/faculty/anderson.html
The Ersatz Brain Project is an attempt to design a suitable computer
for the efficient execution of the software now being developed that
will display human-like cognitive abilities. Examples of these
software applications would be natural language understanding, text
processing, conceptually based internet search, natural human-computer
interfaces, cognitively based data mining, sensor fusion, and image
understanding. Requirements of the proposed software are primary in
shaping our hardware design. We suggest a "cortex-power" massively
parallel computer is technically feasible, requiring on the order of a
million simple CPUs and a terabyte of memory for connections between
CPUs. This approach might build a shoddy and second-rate cortex, but
still perhaps interesting. We will discuss initial "back of the
envelope" ideas about architectures and three possible very early
examples of the unusual software suitable for problems that might run
on such a machine: sensor fusion, simple arithmetic operations, and
one kind of contextual disambiguation.
About the Speaker: Dr. Anderson has been a member of the faculty of
Brown University since September 1973. He is now Professor in the
Department of Cognitive and Linguistic Sciences. He was Chair of the
Department from 1993 to 1998 and in 2000-2001. Dr. Anderson has
published extensively in the area of computational models for
cognition and memory and computational neuroscience. Dr. Anderson is
the author of numerous books and journal articles, including "Talking
Nets," "Introduction to Neural Networks" and "Neurocomputing," Volumes
1 and 2, all from MIT Press. Dr. Anderson has a B.S. in physics
(1962) and a Ph.D. in physiology (1967) from the Massachusetts
Institute of Technology and had postdoctoral fellowships at UCLA
(1967-1971) and Rockefeller University (1971-1973).
____________
PARC FORUM
on Thursday, 13 May 2004, 4:00pm - 5:00pm
George Pake Auditorium, PARC
http://www.parc.com/forum/
"Actor-Oriented Design: Concurrent Models as Programs"
Edward A. Lee
UC Berkeley
Concurrent, domain-specific languages such as Simulink, LabVIEW,
Modelica, VHDL, SystemC, and OPNET provide modularization mechanisms
that are significantly different from those in prevailing
object-oriented languages such as C++ and Java. In these languages,
components are concurrent objects that communicate via messaging,
rather than abstract data structures that interact via procedure
calls. Although the concurrency and communication semantics differ
considerably between languages, they share enough common features that
we consider them to be a family. We call them actor-oriented
languages.
Actor-oriented languages, like object-oriented languages, are about
modularity of software. I will argue that we can adapt for
actor-oriented languages many (if not all) of the innovations of OO
design, including concepts such as the separation of interface from
implementation, strong typing of interfaces, subtyping, classes,
inheritance, and aspects. I will show some preliminary implementations
of these mechanisms in a Berkeley system called Ptolemy II.
About the Speaker: Edward A. Lee is a Professor in the Electrical
Engineering and Computer Science Department at U.C. Berkeley. His
research interests center on design, modeling, and simulation of
distributed, embedded, real-time computational systems. He is a
director of Chess, the Berkeley Center for Hybrid and Embedded
Software Systems, and is the director of the Berkeley Ptolemy
project. He is co-author of five books and numerous papers. His
bachelors degree (B.S.) is from Yale University (1979), his masters
(S.M.) from MIT (1981), and his Ph.D. from U. C. Berkeley (1986). From
1979 to 1982 he was a member of technical staff at Bell Telephone
Laboratories in Holmdel, New Jersey, in the Advanced Data
Communications Laboratory. He is a co-founder of BDTI, Inc., where he
is currently a Senior Technical Advisor, and has consulted for a
number of other companies. He is a Fellow of the IEEE, was an NSF
Presidential Young Investigator, and won the 1997 Frederick Emmons
Terman Award for Engineering Education.
____________
SSP10: SYMBOLIC SYSTEMS FORUM
on Thursday, 13 May 2004, 4:15pm
Bldg. 380:380C (Math Corner)
http://symsys.stanford.edu:8081/ssp-dynamic/servlet/ssp_events
"How people coordinate with each other with and without language"
Herbert H. Clark
Psychology, Stanford
When people engage in joint activities--waltzing, playing tennis,
planning parties, negotiating contracts, or merely conversing--they
have to coordinate on what each of them is to do when and where. They
achieve that coordination not only by means of language, but by means
of what I will call material signals, signals in which they deploy
material objects around them. These actions include pointing at,
placing, and exhibiting objects, but also other actions on and with
objects. I will describe when and how people use both linguistic and
material signals to coordinate in several joint activities.
____________
STANFORD PHONOLOGY WORKSHOP
on Thursday, 13 May 2004, 5:30pm
Margaret Jacks Hall 460:126
http://www-linguistics.stanford.edu/pinterest/
"The interaction of multiple sources of information
in Korean sentence processing"
Hee-Sun Kim
Stanford University
This study presents an investigation of the human sentence processing
mechanism focusing on when and how multiple sources of information are
used in spoken language processing. Two of the most influential but
competing views have been examined: the serial model and the
constraint-based model. The serial model is a syntax-first view in
which syntactic information guides the initial parse of sentences.
Non-syntactic information plays a minor role by entering after the
initial parse is completed and filtering the output from the previous
stage if necessary. The constraint-based model is an interactive
model in which multiple sources of information participate in any
steps of processing as they become available. Explanation is still
needed on exactly how individual informational components contribute
to parsing decision relative to each other where more than two
informational sources interact. This study investigated Korean spoken
sentence processing with the goal of evaluating the two opposing views
on the nature of information access and use in sentence processing.
Specifically, two questions motivated this study: the temporal
relationship of information and the relative contribution of
information in sentence processing. In order to understand the
mechanism, this study explores the use of three types of information:
syntactic, semantic and prosodic information. It uses both on-line and
off-line studies, two distinct but complementary experimental
techniques. In the on-line study, a cross-modal naming task was used
to explore the interaction of the three informational constraints on
the resolution of temporary ambiguity in variously manipulated
semantic and prosodic conditions. We found that the relationship
between informational constraints predicted the speed of processing,
indicating the interaction of information in on-line sentence
processing. In the off-line study, an end-of-sentence comprehension
task was used to investigate how the informational constraints
communicate for the resolution of global ambiguity. The results show
that individual constraints can have different weights in parsing
decision. Taken together, the interaction of information in the
on-line study is in accord with the constraint-based model. The
findings from the off-line study further suggest that the influences
of individual information in different weights should be incorporated
into the constraint-based models of sentence processing.
____________
CS547: HUMAN-COMPUTER INTERACTION SEMINAR
on Friday, 14 May 2004, 12:30-2:00pm
Gates B01
http://www-pcd.stanford.edu/cs547/
"Collaboration, Tool Use, and Work Practice of Mars Mission Scientists"
Roxana Wales and Alonso Vera
NASA Ames
Before the MER rovers landed on Mars, our team spent three years doing
research on and participating in the design of work process and tools
for the mission. Pre-mission influences and tests were the basis for
our assumptions when making recommendations for work process design
and for our design decisions for certain mission tools. The activities
and processes we have observed during the actual mission often did not
match our expectations of what the work would look like. What did we
miss in making our design decisions? Perhaps more importantly, what
did we take as bedrock from our observation of pre-mission events that
turned out to be subject to change during mission operations? Even
with a great deal of user interaction/feedback and in-situ research,
it remains challenging to get things right when designing tools and
processes to a model of what we think events will look like as opposed
to designing tools and processes for on-going operations. We will talk
about the real-time mission drivers that affected and changed the MER
mission. Focusing on the tools and work practice of the daily science
analysis period, we will explore the confounding variables in our
research.
About the Speakers: Roxana Wales is a human-centered computing
Research Scientist with SAIC at the NASA Ames Research Center. She is
an ethnographer who is especially interested in applying the method in
technological settings to reveal explicit as well as implicit
information that is essential for creating better work systems and
supporting technologies within expert domains. She has spent three and
a half years working on the design of surface operations for the Mars
Exploration Rovers Mission at the Jet Propulsion Laboratory in
Pasadena and has worked on projects at NASA's Johnson Space Center
within Mission Control for both the International Space Station and
the Space Shuttle. She also conducted a two and a half year
ethnographic study of airline operations and airline delays at a major
airline. Roxana has a Ph.D. in cultural psychology from the California
Institute of Integral Studies.
Alonso Vera received a PhD in Psychology from Cornell University in
1990 and then held a postdoctoral fellowship in Computer Science at
Carnegie Mellon University until 1993. He worked on human performance
modeling and human-computer interaction with Bonnie John and Allen
Newell. His research has remained focused on those two areas through 6
years as faculty at the University of Hong Kong and 3 and half years
at NASA. Since being at NASA, HCI efforts have been more applied and
targeted toward mission systems, especially for the Mars Exploration
Rovers mission.
____________
NLP READING GROUP
on Friday, 14 May 2004, 2:00pm - 3:30pm
Ventura 17
http://www-nlp.stanford.edu/nlp/nlpgroup.html
"Using Curvature and Markov Clustering in Graphs
for Lexical Acquisition and Word Sense Discrimination"
by Beate Dorow, Dominic Widdows, Katarina Ling,
Jean-Pierre Eckmann, Danilo Sergi, Elisha Moses
Paper available from the NLP reading group web page
Discussion leader: Dominic Widdows
We introduce two different approaches for clustering semantically
similar words. We accommodate ambiguity by allowing a word to belong
to several clusters.
Both methods use a graph-theoretic representation of words and their
paradigmatic relationships. The first approach is based on the
concept of curvature and divides the word graph into classes of
similar words by removing words of low curvature which connect several
dispersed clusters.
The second method, instead of clustering the nodes, clusters the links
in our graph. These contain more specific contextual information than
nodes representing just words. In so doing, we naturally accommodate
ambiguity by allowing multiple class membership.
Both methods are evaluated on a lexical acquisition task, using
clustering to add nouns to the WordNet taxonomy. The most effective
method is link clustering.
____________
CS528: BROAD AREA COLLOQUIUM FOR
AI-GEOMETRY-GRAPHICS-VISION-ROBOTICS
on Monday, 17 May 2004, 4:15pm
TCSeq 200
http://graphics.stanford.edu/courses/cs528/
"Life-Sized Learning"
Leslie Pack Kaelbling
Computer Science and Artificial Intelligence Laboratory (CSAIL), MIT
In the last 10 years, the combination of techniques from machine
learning and operations research has allowed major advances in
learning and planning for uncertain environments. Reasonably large
problems can be solved using current techniques. But what if we want
to scale up to the uncertain learning and planning problem that you
face every day? It is many orders of magnitude larger than the biggest
problem we can solve currently.
In this talk, I'll describe three early pieces of work that try to
begin to address working in truly huge environments. The first is a
method for learning probabilistic rules to describe naive physics
models of the interactions between objects. The second is an uncertain
planning algorithm that uses the rules learned by the first method to
construct contingency plans that consider enough cases to perform
robustly, but are much smaller than complete policies. The last piece
is preliminary work on combining multiple abstraction methods
dynamically, in order to allow an agent to have a working model of the
environment that changes focus depending on the current situation.
About the Speaker: Leslie Pack Kaelbling is Professor of Computer
Science and Engineering and a Research Director of the Computer
Science and Artificial Intelligence Laboratory (CSAIL) at the
Massachusetts Institute of Technology. She has previously held
positions at Brown University, the Artificial Intelligence Center of
SRI International, and at Teleos Research. She received an A. B. in
Philosophy in 1983 and a Ph. D. in Computer Science in 1990, both from
Stanford University. Prof. Kaelbling has done substantial research on
designing situated agents, mobile robotics, reinforcement learning,
and decision-theoretic planning. In 2000, she founded the Journal of
Machine Learning Research, a high-quality journal that is both freely
available electronically as well as published in archival form; she
currently serves as editor-in-chief. Prof. Kaelbling is an NSF
Presidential Faculty Fellow, a former member of the AAAI Executive
Council, the 1997 recipient of the IJCAI Computers and Thought Award,
a trustee of IJCAII and a fellow of the AAAI.
____________
CCRMA COLLOQUIUM
on Wednesday, 19 May 2004, 5:15pm
CCRMA Ballroom, The Knoll
http://www-ccrma.stanford.edu/
"Tabla Transcription"
Parag Chordia
CCRMA
Transcription is a central part of music analysis in traditions where
music is not performed from a score, when no score of a performance is
available or when such a score does not contain the needed
information.
Manual transcription is often very time consuming making large scale
transcription projects tedious and difficult. Tabla transcription of
an average length solo might require notating tens of thousands
strokes.
Up to now, non-heuristic studies of tabla have not been possible
because of the unavailability of sufficient score data. A primary goal
of this project is to create scores that can be used for music
analysis.
The talk will discuss automatic tabla transcription. Like most
automatic music transcription projects the goal is to abstract
musically salient information starting at the signal level. The
current talk will describe the specific problem architecture and the
ways in which general transcription methods have been applied and
extended to solve the sub-problems of event detection, rhythm
detection, stroke (timbre) identification, symbolic representation,
and music typesetting.
____________
SSP10: SYMBOLIC SYSTEMS FORUM
on Thursday, 20 May 2004, 4:15pm
Bldg. 380:380C (Math Corner)
http://symsys.stanford.edu:8081/ssp-dynamic/servlet/ssp_events
"Making sense of it all: Pattern prediction in infancy"
Natasha Kirkham
Psychology Department
We learn about regularly-occurring events in the perceptual
environment by ascertaining associations among temporally adjacent
stimuli in input sequences. Recent research has shown that infants
are remarkably adept at learning the input statistics of auditory,
visual, and multimodal sequences. I will be discussing this research
in terms of a learning mechanism that supports infants' understanding
of their perceptual environment.
____________
STANFORD PHONOLOGY WORKSHOP
on Thursday, 20 May 2004, 5:30pm
Margaret Jacks Hall 460:126
http://www-linguistics.stanford.edu/pinterest/
"The Origin of Algonquian Complex Stem Morphology:
Evidence from Yurok"`
Andrew Garrett
UC Berkeley
Wiyot and Yurok, two indigenous languages of California, are related
to the Algonquian language family. The Yurok verbal category dubbed
"non-inflected" in Robins' grammar seems to be marked by truncation of
inflectional suffixes and part of the stem, but the truncation would
be typologically anomalous if not unique in its phonological
character. Based on a new analysis of Yurok stem-internal morphology
and an examination of the argument structure and discourse function of
these inflectionless verbs, I argue that they are not truncated.
Instead, they are the regular phonological reflexes of an archaic
formation containing only a root plus a classificatory (body-part or
event- or subject-classifying) suffix, a formation originally
construed with an inflecting light verb. This bipartite construction,
I argue, is the ancestor of the morphologically complex verbs found in
Algonquian (and Wiyot, and of Yurok inflected verbs); and such a
construction is at home in the linguistic area where all these
languages originated.
____________
CS547: HUMAN-COMPUTER INTERACTION SEMINAR
on Friday, 21 May 2004, 12:30-2:00pm
Gates B01
http://www-pcd.stanford.edu/cs547/
"Giving Users Control in Ubiquitous Computing Environments"
Anind Dey
Intel Research and UC Berkeley
http://www.cs.berkeley.edu/~dey/
Supporting feedback and control in traditional desktop applications is
necessary to help users understand and interact with these
applications. With implicit data collection, synthesis and action
being key parts of most ubiquitous computing (ubicomp) visions,
feedback and control are even more important in this domain. However,
current ubicomp applications and tools for building them have little
support for providing users with feedback and control.
In this talk, I describe a key subset of ubicomp applications known as
context-aware applications, which adapt to their context of use.
Providing appropriate feedback and control in context-aware systems is
crucial to a "good" user experience. An important aspect of giving end
users control is allowing them to build and evolve their own
context-aware applications. I present two systems that support this,
without requiring any programming by end users. To support feedback in
context-aware systems, I present research in applications and toolkits
that provides information about past, present and future states of an
application.
About the Speaker: Anind Dey is a Senior Researcher at Intel Research
Berkeley and an Adjunct Assistant Professor in the Electrical
Engineering and Computer Science Department at UC Berkeley. He
received a PhD and Masters degree in Computer Science from Georgia
Tech, a Masters degree in Aerospace Engineering from Georgia Tech and
a Bachelors degree in Applied Science (Computer Engineering) from
Simon Fraser University. He performs research at the intersection of
ubiquitous computing (ubicomp) and human-computer interaction,
building tools that make it easier to build useful ubicomp
applications and supporting end users in controlling their ubicomp
systems.
____________
LEARNING, DESIGN & TECHNOLOGY (LDT) MASTERS PRESENTATIONS
on Friday, 21 May 2004, 3:00pm - 6:00pm
Wallenberg Hall Learning theater (Bldg. 160)
http://www.stanford.edu/dept/SUSE/ldt/projects.htm
http://ldt.stanford.edu/
Please join the 2004 Learning, Design & Technology (LDT) masters
students as they present their innovative educational designs and
products! This year the Expo will be held in Wallenberg Hall (Building
160 on the Stanford Quad) from 3:00 to 6:00 PM, next Friday, May 21.
This year we have 10 students presenting 7 projects:
Eric Bailey & Peter Worth
Making Thinking Visible: A technology-mediated instruction plan that
engages 5th through 8th graders in higher-level critical analyses of
audio/video advertisements, television shows, music videos, and films.
Tamecia Jones
Teaching Circuit Theory Using Models & Analogy: An interactive
exhibit being developed for the Tech Museum of Innovation to teach
electric circuit theory using analogies to upper elementary and
middle school students.
Eric Meyers & Cassidy Puckett
Becoming Oakland - Oral Histories of the Emerging West: A web site
that includes oral histories of the Fruitvale community in Oakland
Kihyun Ryoo
Science of Wizardry: Interactive, inquiry-based science software,
designed to help elementary language minority students build
conceptual models of scientific phenomena and articulate their
conceptualizations in academic English
Hillary Thompson & Sandy Johnson
Back Seat Driver: An interactive, portable toy designed to promote
geographical and cultural literacy.
Gregory Umgelter
Language Access at Your Finger Tips (LAFT): A wireless interactive
learning device that will serve as an interpreter of RFID tagged
objects to facilitate language learning.
Paula Wellings
Digital Debrief: A wireless streaming-video system that aggregates
people's real-time and reflective experiences of events through the
use of video and coded responses.
____________
STANFORD SEMANTICS AND PRAGMATICS WORKSHOP
on Friday, 21 May 2004, 3:30pm
Margaret Jacks Hall 460:126
http://www-linguistics.stanford.edu/semgroup/
"Macro-events:
Principles of Event Encoding at the Syntax-Semantics Interface"
Jumlrgen Bohnemeyer
SUNY Buffalo
This presentation examines the principles that govern the encoding of
complex events at the syntax-semantics interface ('correspondence
rules', in the parlance of Jackendoff 1997). The focus will be on the
domain of complex motion events, i.e., events that involve sequences
of location changes of a single Figure with respect to multiple
Grounds (in the terminology of Talmy 2000). Data will be drawn from
research by the author and his colleagues at the Event Representation
project of the Max Planck Institute for Psycholinguistics.
As a starting point for meaningful generalizations about the mapping
of semantic/conceptual event representations into syntax, the
macro-event property (MEP) is proposed (borrowing the term
'macro-event' from Talmy 2000). The MEP applies to constructions that
"package" the parts of an event so tightly as to not permit individual
access by time-positional operators (adverbials, tenses, etc.).
Consider the utterances in (1), which could all serve as descriptions
of the same perceived event:
(1) a. Floyd left Nijmegen at eight. He passed through Utrecht at nine
and reached Amsterdam at ten.
b.*Floyd went from Nijmegen at eight to Amsterdam at ten via Utrecht
at nine.
c. In the morning, Floyd went from Nijmegen to Amsterdam via Utrecht.
Assuming that events as intentional objects of cognitive
representations are individuated by the space-time regions they
occupy, the subevents of departure, passing, and arrival are
individuated in (1a), but not in (1c). Only time-positional adverbials
that denote intervals accommodating all subevents may combine with
(1c). This is what is captured by saying that (1c), but not (1a), has
the MEP.
The MEP is formally defined in the framework of Krifka's (1998)
merological approach to event semantics. It is demonstrated with
reference to multi-verb constructions in a variety of languages that
the MEP does not necessarily apply to a particular type of syntactic
projection (such as the verb phrase or clause), and that it is
difficult to identify a single construction type to which the MEP
applies crosslinguistically. Where mismatches occur, the MEP is a
better predictor of form-to-meaning mapping properties than the type
of syntactic projection.
There is a surprising amount of crosslinguistic variation in how much
information about a motion event can be packaged into an expression
that has the MEP (for short, a 'macro-event expression' (MEE)). Yet, a
number of correspondence rules have been found to hold for MEEs in all
languages examined so far. One such rule requires biunique assignment
of thematic relations (Bresnan 1982, Chomsky 1981, Fillmore 1968) to
arguments and adjuncts within MEEs, including to Ground phrases in
motion event descriptions. That this principle appears to operate
specifically on MEEs confirms Carlson's (1984, 1998) hypothesis
according to which biunique linking is conceptually tied to the
individuation criteria for events. Time permitting, a variety of
putative correspondence rules on MEEs will be discussed, including one
domain-specific constraint on motion event descriptions, the 'Unique
Vector Constraint' (Bohnemeyer 2003).
____________
STANFORD CONFERENCE
on Friday, Saturday, and Sunday, 21-23 May 2004
Stanford University
Sponsored by the Division of Literatures, Cultures and Languages
http://dlcl.stanford.edu/research/workgroups/diversity.html
"Diversity and Universals in Language: The Consequences of Variation"
Conference Program
Friday, 21 May 2004 (Bldg. 260 - Rm. 113)
7:00pm Welcome
7:30pm Marianne Mithun (University of California, Santa Barbara)
"Divergence and confluence: typology, diachrony, and contact"
8:30pm Reception
Saturday, 22 May 2004 (Bldg. 300 - Rm. 300T)
9:00am Nikolaus Ritt (University of Vienna)
"A Darwinian perspective on languages, varieties, and
universals"
9:30am Hiromi Ozeki (University of Tokyo) and Yasuhiro Shirai
(Cornell University)
"The consequences of variation in the acquisition of relative
clauses: An analysis of longitudinal production data from five
Japanese children"
10:00-10:15 Break
10:15am Reijirou Shibasaki (University of California, Santa Barbara)
"Explorations of noun-modifying tautological constructions
across languages: with special reference to X to-yuu X in
Japanese"
10:45am Jennifer Mittelstaedt (Georgetown University)
"Apparent-time change in the Smith Island Auxiliary Verb System"
11:15-11:30 Break
11:30am Barbara Johnstone (Carnegie Mellon University)
"Three Ways To Sound Like a Pittsburgher: Stancetaking and
Vernacular Norm-Formation"
12:30-2:15 Lunch on campus including a presentation by the Stanford
Japanese Dialect Research Group.
2:15pm John Beavers, Beth Levin, and Shiao-Wei Tham
(Stanford University)
"A morphosyntactic basis for variation in the encoding of
motion events"
2:45pm Elena Maslova (Stanford University)
"Cross-linguistic and language-internal variation as a
manifestation of language universals: the case of
reflexive/reciprocal polysemy"
3:15-3:30 Break
3:30pm Anne-Marie Hartenstein (Rice University)
"The middle voice construction in Romanian -
a corpus based analysis"
4:00pm Mark Donohue (National University of Singapore)
"Voice varieties in Indonesian/Malay"
4:30-4:45 Break
4:45pm Toshio Ohori (Tokyo University)
tba
5:45pm End of first day; Dinner
Sunday, 23 May 2004 (Bldg. 300 - Rm. 300T)
9:00am Prashant Pardeshi, Kaoru Horie, and Qing-Mei Li
(Tohoku University)
"Being on the receiving end: A tour into linguistic variation
at propositional level"
9:30am Jared Bernstein (Ordinate Corporation and Stanford University)
"Workable models of standard performance in English and Spanish"
10:00-10:15 Break
10:15am Jim Miller (University of Auckland)
"Unplanned spoken English: standard or non-standard? clause
syntax or discourse organisation?"
10:45am Yumiko Nishi and Yasuhiro Shirai (Cornell University)
"Where L1 semantic transfer occurs: The significance of
cross-linguistic variation in lexical aspect in the universal
phenomena of L2 aspect acquisition"
11:15-11:30 Break
11:30am Claire Kramsch (University of California, Berkeley)
tba
12:30pm Conference ends
____________
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 (i.e., no abstracts). 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/Archive/calendar/current.shtml
and the archives at
http://www-csli.stanford.edu/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://news.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
____________