[Date Prev][Date Next][Date Index]
CSLI Calendar, Wednesday, 28 November 2007, vol. 23:13
CSLI CALENDAR OF PUBLIC EVENTS
______________________________________________________________________
28 NOVEMBER 2007 Stanford Vol. 23, No. 13
______________________________________________________________________
A weekly publication of the
Center for the Study of Language and Information (CSLI)
a subdivision of H-STAR, http://hstar.stanford.edu/
Stanford University, Cordura Hall, Stanford, CA 94305-4101
http://www-csli.stanford.edu/
____________
ACTIVITIES FROM 28 NOVEMBER 2007 TO 7 DECEMBER 2007
WEDNESDAY, 28 NOVEMBER 2007
12 noon UC Berkeley IPSR colloquium [28-Nov-07]
5101 Tolman Hall (Berkeley)
"Wired for Good"
Dacher Keltner
UC Berkeley
http://ls.berkeley.edu/dept/ipsr/colloquia.html
12:15pm Psychology Developmental Brownbags [28-Nov-07]
Jordan Hall 420:102
Title to be announced
Anne Fernald
Stanford University
http://www-psych.stanford.edu/events_developmental.html
1:00pm Berkeley talk [28-Nov-07]
Soda Hall 405 (UC Berkeley)
"The Next Generation of Neural Networks"
Geoffrey Hinton
Computer Science, University of Toronto
Abstract below
3:15pm Stanford Phonology Workshop [28-Nov-07]
Linguistics Chair's office, Margaret Jacks Hall
"Rethinking linguistic models of rhythm
through evidence from jazz bop swing"
Stephanie Shih
Stanford University
http://www-linguistics.stanford.edu/pinterest/
Abstract below
3:45pm Psychology Department Colloquium [28-Nov-07]
Jordan Hall 420:041
Title to be announced
Alvaro Pascual-Leone
Harvard University
http://www-psych.stanford.edu/events_colloquium.html
4:15pm EE380: Computer Systems Lab Colloquium [28-Nov-07]
Gates B03 (NEC Auditorium)
"The Google G-Phone"
speaker to be announced
http://ee380.stanford.edu/contents.html
THURSDAY, 29 NOVEMBER 2007
10:00am Academic Freedom and the Ethics of Research [29-Nov-07]
Tresidder, Oak West
"The Structure of Academic Freedom"
Robert Post
Yale Law School
http://ethicsinsociety.stanford.edu/lectures.html
11:00am Academic Freedom and the Ethics of Research [29-Nov-07]
Tresidder, Oak West
"Academic Freedom After 9/11"
John Etchemendy, Provost at Stanford
Stephen Monismith, Stanford, Civil and Environmental Engineering
http://ethicsinsociety.stanford.edu/lectures.html
12 noon CSLI CogLunch [29-Nov-07]
Cordura Hall 100
"Meaningful Talk"
Yossi Feinberg
GSB, Stanford
http://www-csli.stanford.edu/events/Coglunch/
Abstract below
12:15pm Stanford Networking Seminar [29-Nov-07]
Packard 101
"Embracing Disruptive Technologies
such as P2P Technology and Community Networks"
Anja Feldmann
Deutsche Telekom Laboratories/TU Berlin
http://netseminar.stanford.edu/
1:00pm UC Berkeley Trust Seminar [29-Nov-07]
Wozniak lounge, Soda Hall
"Building Reliable Voting Machine Software"
Ka-Ping Yee
UC Berkeley
http://www.truststc.org/seminar.htm
Abstract below
1:30pm Academic Freedom and the Ethics of Research [29-Nov-07]
Tresidder, Oak West
"Academic Freedom and Commerce"
Drummond Rennie, UCSF, Health Policy / JAMA, Deputy Editor
Joseph Bankman, Stanford Law School
http://ethicsinsociety.stanford.edu/lectures.html
4:00pm PARC Forum [29-Nov-07]
George Pake Auditorium at PARC
"StumbleUpon: A Web Discovery System"
Garrett Camp
Stumble Upon
http://www.parc.com/forum/
Abstract below
4:00pm UC Berkeley CIS Seminar [29-Nov-07]
Soda Hall 320 (UC Berkeley)
"Consistency of the group Lasso and multiple kernel learning"
Francis Bach
Ecole Normale Superieure / INRIA
http://www.cs.berkeley.edu/~rakhlin/cis-seminar
4:00pm UC Berkeley Talk [29-Nov-07]
Wang Room, 531 Cory (UC Berkeley)
"Computationally-based Agnostic Inference"
Terrence L. Fine
Electrical and Computer Engineering, Cornell University
http://www.eecs.berkeley.edu/cal/
Abstract below
4:15pm SSP10: Symbolic Systems Forum [29-Nov-07]
Bldg. 380:380C (Math Corner)
"How Science Thinks:
The Science and Engineering of Science and Engineering"
Jeff Shrager
Symbolic Systems
http://symsys.stanford.edu/ssp_events
Abstract below
4:15pm Information Systems Seminar [29-Nov-07]
Packard 101
"Statistical Analysis of Online News"
Laurent El Ghaoui
Electrical Engineering and Computer Science, UC Berkeley
http://isl.stanford.edu/colloquium.html
Abstract below
4:15pm Fundamental Themes in Neuroscience Seminar [29-Nov-07]
Munzer Auditorium, Beckman
"From synapse to regeneration"
Yishi Jin
Biological Sciences, Howard Hughes Medical Institute
http://www.biology.ucsd.edu/labs/yishijin/
http://nis-seminars.stanford.edu/
FRIDAY, 30 NOVEMBER 2007
12:30pm CS547: Human-Computer Interaction Seminar [30-Nov-07]
Gates B01
"Context Aware Computing:
Understanding and Responding to Human Intention"
Ted Selker
MIT Media Lab
http://web.media.mit.edu/~selker/
http://hci.stanford.edu/cs547/
Abstract below
12:30pm UC Berkeley HWNI Talk [30-Nov-07]
101 LSA (Berkeley)
"Neural mechanisms of action learning: From intent to habit"
Rui Costa
NIH
http://neuroscience.berkeley.edu/events/
3:00pm Berkeley Information Access Seminar [30-Nov-07]
107 South Hall (Berkeley)
"Terminology Mapping for Distributed Search"
Vivien Petras
GESIS-IZ, Bonn, Germany
"Conceptual Schemas for Events: Final Progress Report"
Ryan Shaw
http://courses.ischool.berkeley.edu/i296a-1/f07/schedule.html
Abstract below
3:30pm Psychology/Linguistics/Philosophy Colloquium [30-Nov-07]
Jordan Hall 420:041
"Assessing the relationship between Piraha, language,
cognition, and culture: What is at stake and how to proceed"
Ted Gibson
MIT
http://tedlab.mit.edu/
http://www-psych.stanford.edu/events_cognitive.html
Abstract below
4:10pm UC Berkeley Logic and the Methodology of Science [30-Nov-07]
60 Evans Hall (Berkeley)
"Will-o'-the-Wisp?: In Pursuit of a Foundation for
Unrestricted Category Theory"
Solomon Feferman
Stanford University
http://logic.berkeley.edu/colloquium.html
4:30pm UC Santa Cruz Linguistics Colloquium [30-Nov-07]
Silverman Conference Room, Stevenson College (UC Santa Cruz)
"The Role of Prosody in Constituent Ordering"
Arto Anttila
Stanford University
http://ling.ucsc.edu/news_events/colloquia/
SATURDAY, 1 DECEMBER 2007
4:00pm Big Game
Stanford football stadium
"Will Cal make it 6 in a row"
football team, Stanford
football team, UC Berkeley
MONDAY, 3 DECEMBER 2007
12 noon Wallenberg Brown Bag [3-Dec-07]
Wallenberg Hall
Chris Thomsen
IRiSS
12 noon UC Berkeley HWNI Talk [3-Dec-07]
5101 Tolman Hall (Berkeley)
"Plasticity and Learning in the Rodent Visual System"
Natalia Caporale
Thesis Seminar, Dan Lab
http://neuroscience.berkeley.edu/events/
12:45pm Center for Internet and Society Talk [3-Dec-07]
Law School 280B
"Faceoff: Lessig v. Zittrain"
Lawrence Lessig, Stanford Law
Jonathan Zittrain, Oxford Law
http://cyberlaw.stanford.edu/
2:00pm UC Berkeley Working Group in the Philosophy of Mind [3-Dec-07]
5101 Tolman Hall (Berkeley)
"Reading and discussion of: Jerry Fodor, 'The Mind Doesn't
Work That Way', chaps. 1,2 & 3"
http://neurophilosophy.berkeley.edu/meetings.html
4:00pm UC Berkeley Ear Club [3-Dec-07]
3105 Tolman Hall (Berkeley)
"A role of cognitive effort when extracting speech from noise"
Tassos Sarampalis
Psychology, UC Berkeley
http://ear.berkeley.edu/ear-club-schedule.html
4:10pm UC Berkeley Philosophy Department Colloquium [3-Dec-07]
Howison Philosophy Library, (305 Moses Hall) (Berkeley)
"Perceptual Experience and Error"
James Genone
UC Berkeley
http://philosophy.berkeley.edu/
6:00pm Computer Musings [3-Dec-07]
Skilling Auditorium
"Sideways Heaps: The Thirteenth Annual Christmas Tree Lecture"
Don Knuth
http://www-cs-faculty.stanford.edu/~knuth/musings.html
TUESDAY, 4 DECEMBER 2007
4:00pm SRI AI Seminar Series [4-Dec-07]
EJ228, SRI International
"NeoVictorian Computing"
Mark Bernstein
Eastgate Systems Inc.
http://www.ai.sri.com/seminars/
4:15pm Mathematical Logic Seminar [4-Dec-07]
Bldg. 420:048
"The Tree of Knowledge in Action"
Eric Pacuit
Stanford
http://www-logic.stanford.edu/
Abstract below
4:30pm Stanford Security Seminar [4-Dec-07]
Gates 4B center area (opposite 490)
"Tradeoffs in Retrofitting Security: An Experience Report"
Mark S. Miller
Google
http://theory.stanford.edu/seclab/sem.html
WEDNESDAY, 5 DECEMBER 2007
12:15pm Psychology Developmental Brownbags [5-Dec-07]
Jordan Hall 420:102
Title to be announced
Linda Smith
Indiana Univ
http://www-psych.stanford.edu/events_developmental.html
3:30pm Bio-X: Talks in English [5-Dec-07]
Clark Center S360
"Neural 'learning rules' for tuning motor programs"
Jennifer Raymond
Neurobiology
http://biox.stanford.edu/news/tie.html
3:45pm Psychology Department Colloquium [5-Dec-07]
Jordan Hall 420:041
"Action, Objects, Shape and Early Word Learning"
Linda Smith,
Indiana University
http://www-psych.stanford.edu/events_colloquium.html
4:15pm EE380: Computer Systems Lab Colloquium [5-Dec-07]
Gates B03 (NEC Auditorium)
"Science communication, science literacy and public support:
new models in place of old thoughts"
Dennis Bartels, Rob Semper
San Francisco Exploratorium
http://www.exploratorium.edu/
http://ee380.stanford.edu/contents.html
4:15pm CREEES Talk [5-Dec-07]
Bolivar House Conference Room, 582 Alvarado
"When Languages Die:
Tracking Global and Local Trends of Language Extinction"
K. David Harrison
Linguistics, Swarthmore College
Research Director, Living Tongues Institute
http://CREEES.stanford.edu/events/
THURSDAY, 6 DECEMBER 2007
10:30am Ontolog Invited Speaker [6-Dec-07]
online/conference call
"If we build it, will they come? Social-engineering of
new technology to disseminate biomedical ontologies"
Mark Musen
Stanford University
http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2007_12_06
Abstract below
12 noon Work, Technology and Organization Colloquium [6-Dec-07]
Terman 217
Title to be announced
Hayagreeva Rao
GSB, Stanford University
http://faculty-gsb.stanford.edu/rao/research.html
(Joint WTO-STVP Colloquium)
http://stvp.stanford.edu/research/seminars.html
http://www.stanford.edu/group/WTO/
12:15pm Stanford Networking Seminar [6-Dec-07]
Packard 101
"Networks on the Edge:
Challenges and Opportunities in Residential Wireless Networks"
Srinivasan Seshan
Carnegie Mellon University
http://netseminar.stanford.edu/
1:00pm UC Berkeley Trust Seminar [6-Dec-07]
Wozniak lounge, Soda Hall
"Quantifying Strengths and Risk Assessments of Software
Protections"
George Cybenko
Dartmouth College
http://www.truststc.org/seminar.htm
4:00pm PARC Forum [6-Dec-07]
George Pake Auditorium at PARC
Title to be announced
Charlene Li
Forrester Research
http://www.forrester.com/rb/analyst/charlene_li
http://www.parc.com/forum/
4:15pm SSP10: Symbolic Systems Forum [6-Dec-07]
Bldg. 380:380C (Math Corner)
"Learning to Recognize Facial Emotions:
Psychologists vs. Artists"
David C. Wilkins
Symbolic Systems, Stanford
http://symsys.stanford.edu/ssp_events
Abstract below
4:15pm Information Systems Seminar [6-Dec-07]
Packard 101
"Exchanging Limits: Why Iterative Decoding Works"
Rüdiger Urbanke
School of Computer & Communication Sciences, EPFL
http://isl.stanford.edu/colloquium.html
Abstract below
4:15pm Fundamental Themes in Neuroscience Seminar [6-Dec-07]
Munzer Auditorium, Beckman
"Imaging circuit development in the vertebrate retina"
Rachel Wong
Washington University
http://wonglab.biostr.washington.edu/
http://nis-seminars.stanford.edu/
FRIDAY, 7 DECEMBER 2007
12:30pm CS547: Human-Computer Interaction Seminar [7-Dec-07]
Gates B01
To be announced
http://hci.stanford.edu/cs547/
12:30pm UC Berkeley HWNI Student Seminar [7-Dec-07]
101 LSA (Berkeley)
"Hippocampal circuits and dynamic range"
Massimo Scanziani
UCSD
http://neuroscience.berkeley.edu/events/
3:00pm Berkeley Information Access Seminar [7-Dec-07]
107 South Hall (Berkeley)
"The Fall in Review"
Clifford Lynch
http://courses.ischool.berkeley.edu/i296a-1/f07/schedule.html
3:15pm Friday Cognitive Seminar [7-Dec-07]
Jordan Hall 420:050
Title to be announced
Yoichiro Masuda
Jikei University School of Medicine (Tokyo) and Stanford University
http://www-psych.stanford.edu/events_cognitive.html
____________
Stanford Blood Center: Shortage of O+, O-, A-, B-, and AB-. For an
appointment: http://bloodcenter.stanford.edu/ or call 650-723-7831.
It only takes an hour of your time and you get free cookies.
____________
NOTES
Best wishes to both Cal and Stanford on this week's competition.
Fencing, Friday, 30 November, 6pm, Burnham Pavilion.
____________
BERKELEY TALK
on Wednesday, 28 November 2007, 1:00pm
Soda Hall 405 (UC Berkeley)
"The Next Generation of Neural Networks"
Geoffrey Hinton
Computer Science, University of Toronto
In the 1980's, new learning algorithms for neural networks promised to
solve difficult classification tasks, like speech or object
recognition, by learning many layers of non-linear features. The
results were disappointing for two reasons: There was never enough
labeled data to learn millions of complicated features and the
learning was much too slow in deep neural networks with many layers of
features. These problems can now be overcome by learning one layer of
features at a time and by changing the goal of learning. Instead of
trying to predict the labels, the learning algorithm tries to create a
generative model that produces data which looks just like the
unlabeled training data. These new neural networks outperform other
machine learning methods when labeled data is scarce but unlabeled
data is plentiful.
____________
STANFORD PHONOLOGY WORKSHOP
on Wednesday, 28 November 2007, 3:15pm
Linguistics Chair's office, Margaret Jacks Hall
http://www-linguistics.stanford.edu/pinterest/
"Rethinking linguistic models of rhythm
through evidence from jazz bop swing"
Stephanie Shih
Stanford University
Background to the research
Jackendoff and Lerdahl's (1983; henceforth, GTTM) grid-based theory of
modeling musical intuitions of rhythm?based on Western classical
music?has formed a foundation for current research on both musical and
linguistic rhythm.
Interdisciplinary issues
Influenced by approaches in describing linguistic rhythm, which have
developed since the 1980s, the conclusions of this research at once
engages in the decades-old debates of metrical theory as well as
explores the limitations of linguistic theory for musical description.
Issue/hypothesis under investigation
When extended to non-classical forms, the GTTM grid theory is too
limited to accurately model and fully capture musical intuitions
governing the cognition of rhythm. From evidence in jazz bop swing, a
form whose rhythm is perceptibly distinct from that of classical
music, this paper develops revisions on the current generative theory
for describing musical rhythm.
Findings/description
Using data from Ella Fitzgerald recordings, I generate a metrical
model of jazz rhythm that departs from GTTM's model in three
significant ways. First, the polyphonic texture of jazz necessitates
the existence of two metrical models which together capture jazz's
rhythmic form, though these models are similar in many ways and may be
reconciled as one. Second, evidence from jazz supports an argument
that musical rhythm, like linguistic rhythm, involves the metrical
features of culminativity and constituency. This argument leads to
the use of a tree-based model for representing rhythm as opposed to
GTTM's grid-based model, which fails to demonstrate constituency
structure. Third, unlike the binary metrical model hitherto used to
describe musical rhythm, jazz rhythm requires ternary metrical
structure, like anapestic meters in poetics, accounting for both the
seemingly uneven rhythm of swing and jazz's conflict between stressing
phrasal beginnings and metrical prominent beats.
Conclusions/future directions
This revised model, based on jazz, accounts for more of the metrical
relationships present in music that possible in GTTM, permitting more
flexibility and comprehensiveness in describing listeners' intuitions
of musical rhythm. Furthermore, the model presented here more greatly
reflects models of linguistic rhythm, a significant conclusion in
regards to the similarities between language and music.
____________
CSLI COGLUNCH
on Thursday, 29 November 2007, 12 noon - 1:00pm
Cordura Hall 100
http://www-csli.stanford.edu/events/Coglunch/
"Meaningful Talk"
Yossi Feinberg
GSB, Stanford
Recent developments in the interface of Economics and Logic yield the
promise of capturing phenomena in strategic interaction that was
previously beyond theoretical economic modeling. We consider one such
application in the case of strategic situations that players may
perceive differently. We show that the content of messages sent in
this setting carries strategic implications without resorting to prior
conventions and beyond the observations made in the literature dealing
with cheap talk communications. The content of the message becomes
strategically meaningful since it reveals perceptions. Various forms
of interaction between meaningful statements and strategic behavior
are discussed.
____________
UC BERKELEY TRUST SEMINAR
on Thursday, 29 November 2007, 1:00pm
Wozniak lounge, Soda Hall (Berkeley)
http://www.truststc.org/seminar.htm
"Building Reliable Voting Machine Software"
Ka-Ping Yee
UC Berkeley
The democracy upon which our modern society is built ultimately
depends on a system that collects and counts votes. In the United
States today, and to an increasing extent elsewhere, nearly every part
of that system relies on computer software in some way. Widely
reported failures in the usability, security, stability, and
correctness of such software have led to a crisis in confidence. I
will discuss ways to achieve confidence in the voting system as a
whole and voting machine software in particular, with emphasis on that
most thorny of software security challenges, the insider attack. How
can we design reliable software, and if someone else designs it, how
can we tell if it is reliable? I will explain why the software in the
voting machine is the most crucial of all, propose a design for
software that is hundreds of times smaller and simpler than that used
in some of today's leading voting machines, and argue that this can
help lead to voting machines that are more reviewable, usable,
accessible, and secure.
About the Speaker: Ka-Ping Yee is a Ph. D. candidate in Computer
Science at UC Berkeley. His graduate research has focused on security
and usability. He participated in this past summer's Voting Systems
Review for the California Secretary of State as a reviewer of voting
system source code, and his work on voting systems has been published
at the USENIX/ACCURATE Electronic Voting Technology workshop.
____________
PARC FORUM
on Thursday, 29 November 2007, 4:00pm - 5:00pm
George Pake Auditorium, PARC
http://www.parc.com/forum/
"StumbleUpon: A Web Discovery System"
Garrett Camp
Stumble Upon
StumbleUpon is a web content discovery service with 4 million
members. It uses a personalized recommendation engine and toolbar
interface to suggest topically relevant web content, such as websites,
photos, or videos recommended by friends or like-minded people. This
lets users ("stumblers") discover great content that they wouldn't
think to search for, and provides a unique platform for distributing
web media. This talk will go though StumbleUpon's approach to online
discovery and the history of its development.
About the Speaker: Garrett Camp, Founder & Chief Product Officer.
Garrett is responsible for StumbleUpon's product design and
strategy. He has guided StumbleUpon's design and development since
2001, from inception to over 3.9M members. Garrett completed his
Masters in Software Engineering at the University of Calgary, where he
researched collaborative systems, evolutionary algorithms and
information retrieval. He is also a 2007 recipient of MIT Technology
Review's TR35 award.
____________
UC BERKELEY TALK
Wang Room, 531 Cory (UC Berkeley)
on Thursday, 29 November 2007, 4:00pm
http://www.eecs.berkeley.edu/cal/
"Computationally-based Agnostic Inference"
Terrence L. Fine
Electrical and Computer Engineering, Cornell University
We address the classical problem of inductive inference, from the
known (evidence) to the unknown (hypothesis). Let h|e denote the
inferential, or inductive, support lent to hypothesis h by evidence e,
both expressed in some language. To Bayesians this is a solved
problem; it is "simply" a matter of determining a conditional
probability P(h|e) of the hypothesis given the evidence from a prior
over all statements in the assumed language. We maintain that the
breadth of application of inductive inference exceeds the Bayesian
scope.
Our starting point is to assume that the original general inference
problem has been recast in a formal language that has as syntax the
set of nite-length binary strings with an allowed operation of
concatenation. The associated semantics is the very basic one of what
can be computed from what using a universal Turing machine that models
a reasoner. Our claim to "agnostic" inference means that we intend to
be careful to minimize our a priori assumptions. We do not believe
that such assumptions can be altogether eliminated. However, we will
make no distributional assumptions.
Our approach is motivated by mid-1960s work of Ray Solomonoff on
universal semimeasures and by a 1921 essay by John Maynard Keynes.
Our goal is not a numerical assessment of individual h|e, be it
conditional probability or otherwise. Instead, we will construct a
partial ordering on pairs h|e and h'|e'. We will then examine the
robustness of this ordering with respect to the choice of Turing
machine or reasoner and sketch applications for the resulting partial
ordering of inferential support.
____________
SSP10: SYMBOLIC SYSTEMS FORUM
on Thursday, 29 November 2007, 4:15pm
Bldg. 380:380C (Math Corner)
http://symsys.stanford.edu/ssp_events
"How Science Thinks:
The Science and Engineering of Science and Engineering"
Jeff Shrager
Symbolic Systems
For over three decades cognitive scientists have been studying how
science works and how scientists think. What have we learned about
scientific cognition and about science as a human activity? How has
this informed cognitive science more generally? How has it helped us
build semi-automated discovery systems and better tools to support
scientific practice and facilitate discovery? How does this all play
with the Web 24.0 vision? (**) In this talk I'll use some of my own,
and a lot of other people's research to lead a guided tour to some
partial answers to these interesting question.
About the Speaker: Jeff Shrager is consulting associate professor of
Symbolic Systems. His work spans human and machine learning and
development, and both computational and "wet" marine biology and drug
discovery. He current leads the Health Care Initiative at CommerceNet
which is using Web 24.0 technology (**) to build Virtual
Pharmaceutical Companies to address rare and orphan diseases.
(** If Web 1.0 is the current web, Web 2.0 the social web, Web 3.0 the
semantic web, and Web 4.0 the programmable web, then Web 24.0
(1*2*3*4) is be the programmable social semantic web. I just made this
term up for this talk, but it's actually rather appropriate, as you'll
see!)
____________
INFORMATION SYSTEMS SEMINAR
on Thursday, 29 November 2007, 4:15pm-5:15pm
Packard 101
http://isl.stanford.edu/colloquium.html
"Statistical Analysis of Online News"
Laurent El Ghaoui
Electrical Engineering and Computer Science, UC Berkeley
Each day we are inundated with an avalanche of online news. Yet is is
currently hard to obtain a global view of this information. What are
the images that various news media project about specific topics, such
as global warming, human rights or presidential candidates? How do
these images evolve over time? How do they differ across different
media sources, scientific or mainstream? What are the dynamics of news
events across news networks?
Modern statistical learning and optimization methods are having a
great impact in fields where large amounts of data have become
recently available, such as biology or finance. With no doubt, such
methods can help shed light on the issues above as well, to the
benefit of the social scientist or the ordinary citizen. In turn,
online news analysis pushes the boundaries of statistics and
optimization towards databases, networks, visualization, and calls for
a renewed interaction between computer engineering and social
sciences.
I will describe a project which aims at providing user-friendly tools
for analyzing large amounts of text data residing in online databases,
with a focus on online news data and voting records. I will discuss in
particular how online learning and sparsity-inducing methods arise as
key ingredients, and I will delineate some related fundamental
challenges.
About the Speaker: Laurent El Ghaoui graduated from Ecole
Polytechnique (Palaiseau, France) in 1985, and obtained his PhD in
Aeronautics and Astronautics at Stanford University in March 1990. He
was a faculty member of the Ecole Nationale Supérieure de Techniques
Avancées (Paris, France) from 1992 until 1999, and held part-time
teaching appointments at Ecole Polytechnique within the Applied
Mathematics department and Université de Paris-I (La Sorbonne) in the
Mathematics in Economy program. In 1998, he was awarded the Bronze
Medal for Engineering Sciences, from the Centre National de la
Recherche Scientifique, France. He joined the Berkeley faculty in
April 1999 as an Acting Associate Professor, and obtained his tenure
in May 2001. He went on leave from UC between July 2003 and September
2006 to work for SAC Capital Management, a hedge fund based in New
York and Connecticut.
____________
CS547: HUMAN-COMPUTER INTERACTION SEMINAR
on Friday, 30 November 2007, 12:30-2:00pm
Gates B01
http://hci.stanford.edu/cs547/
"Context Aware Computing:
Understanding and Responding to Human Intention"
Ted Selker
MIT Media Lab
http://web.media.mit.edu/~selker/
This talk will demonstrate that Artificial intelligence can
competently Improve human interaction with systems and even each other
in a myriad of natural scenarios.
Humans work to understand and react to each others intentions. The
context aware computing group at the MIT Media lab has demonstrated
that across most aspects of our life, computers can do this too. The
groups demonstrations range from car to office kitchen to and even
bed. The goal is to show that human intentions can be recognized
considered and responded to appropriately by computer systems.
Understanding and acting appropriately to intentions requires more
than good sensors, it requires understanding of the value of the
input. The context aware demonstrations therefore rely completely on
models of what the system can do, what the tasks are that can be
performed and what is known about the user . These models of system
task and user form a central basis for deciding when and how to
respond in a specific situation
About the Speaker: Dr. Ted Selker is an Associate Professor at the MIT
Media Laboratory, the Director of the Context Aware Computing Lab,
co-director of the Caltech/MIT Voting Technology Project, and also of
the Counter Intelligence/ Design Intelligence special interest group
on product design of the future. His work strives to demonstrate that
people's intentions can be recognized and respected by the things we
design. His work is recognized for creating demonstrations of a world
in which people's demonstration of desires causes computers to help
them across natural and complex domains, such as kitchens, cars, email
and voting. This work uses sensors and artificial intelligence in
adaptive models of users' systems and tasks to create keyboardless
computers. Ted's work takes the form of prototype concept products
supported by cognitive science research. He particularly works to show
how this approach helps product design to bridge communication gaps
for technology and people. Ted's work is also applied to developing
and testing user experience technology and security architectures for
recording voter intentions securely and accurately. Prior to joining
MIT faculty in November 1999, Ted was an IBM fellow and directed the
User Systems Ergonomics Research lab. He has served as a consulting
professor at Stanford University, taught at Hampshire, University of
Massachusetts at Amherst and Brown Universities and worked at Xerox
PARC and Atari Research Labs.
Ted's research has contributed to products ranging from notebook
computers to operating systems. He is known for the design of the
TrackPoint in-keyboard pointing device found in many notebook
computers, as well as many other innovations at IBM. Ted's work has
resulted in numerous awards, patents, and papers and is often featured
by the press. Ted was co-recipient of the Computer Science Policy
Leader Award for Scientific American 50 in 2004 and the American
Association For People with Disabilities Thomas Paine Award for his
work on voting technology in 2006.
____________
BERKELEY INFORMATION ACCESS SEMINAR
on Friday, 30 November 2007, 3:00pm - 5:00pm
107 South Hall (Berkeley)
http://courses.ischool.berkeley.edu/i296a-1/f07/schedule.html
"Terminology Mapping for Distributed Search"
Vivien Petras
GESIS-IZ, Bonn, Germany
Between 2004-2007, the German Federal Ministry for Education and
Research funded a major terminology mapping initiative at the GESIS
Social Science Information Centre in Bonn (GESIS-IZ) with the task to
organize, create and manage 'cross-concordances' between major
controlled vocabularies (thesauri, classification systems, subject
heading lists) centred around the social sciences but quickly
extending to other subject areas. To date, 62 mappings between 25
different controlled vocabularies and almost half a million relations
were created. I will introduce the project, show some possible
applications and present results from an evaluation effort that was
targeted toward measuring the effectiveness of these mappings in
search.
"Conceptual Schemas for Events: Final Progress Report"
Ryan Shaw
This fall I looked at theories or models of event structure and event
relationships developed by philosophers, cognitive psychologists, and
researchers in the multimedia computing and Semantic Web
communities. I will present some conclusions about these models and
their usefulness as a basis for designing systems that index documents
in terms of historical events. I will end by speculating on some
alternative approaches that may be more useful and discuss upcoming
work to implement these approaches in a working prototype.
____________
PHILOSOPHY/PSYCHOLOGY/LINGUISTICS DEPARTMENT COLLOQUIUM
on Friday, 30 November 2007, 3:15pm
Jordan Hall 420:041
http://www-philosophy.stanford.edu/ce.html
"Assessing the relationship between Piraha, language, cognition, and culture:
What is at stake and how to proceed"
Ted Gibson
MIT
http://tedlab.mit.edu/
A foundational assumption of many researchers investigating the
universals of human language is that many properties of language are
independent of the cultural context and the non-linguistic cognitive
abilities of the(ir) speakers. But it's not clear that this
assumption is warranted. Everett (2005) described the case of the
Piraha, an isolated Amazonian tribe who are allegedly characterized
by very unusual linguistic and non-linguistic cognitive properties
(e.g., finite language, lack of words for numbers and colors, lack of
quantifiers). Critically, he argued that all these properties can be
accounted for by a general cultural constraint against abstraction.
The validity of these claims remains an open question. I will report
some initial results from a set of experiments I conducted in
collaboration with Mike Frank and Ev Fedorenko during a visit to a
Piraha village in January 2007 in order to test some of these claims.
____________
MATHEMATICAL LOGIC SEMINAR
on Tuesday, 4 December 2007, 4:15pm
Bldg. 420:048
http://www-logic.stanford.edu/
"The Tree of Knowledge in Action"
Eric Pacuit
Stanford
Many logical systems today describe intelligent interacting agents
over time. Frameworks include Interpreted Systems (IS),
Epistemic-Temporal Logic (ETL), and Dynamic Epistemic Logic
(DEL). This variety is an asset, as different modeling tools can be
fine-tuned to specific applications. But it may also be an obstacle,
when barriers between paradigms and schools go up. In this talk, I
will discuss some of the technical issues that arise when comparing
these systems. More specifically, I will
1. Survey a number of decidability and undecidability results
concerning Epistemic Temporal Logic. The goal is to provide a general
picture drawing on ideas from different areas of modal logic.
2. Rigorously compare Dynamic Epistemic Logic and Epistemic Temporal
Logic.
____________
ONTOLOG INVITED SPEAKER
on Thursday, 6 December 2007, 10:30am PST
online/conference call
http://ontolog.cim3.net/
"If we build it, will they come? Social-engineering of
new technology to disseminate biomedical ontologies"
Mark Musen
Center for Biomedical Informatics Research, Stanford University
RSVP requested to peter.yim...cim3.com
The National Center for Biomedical Ontology is developing a wide range
of Web-based services to assist end users in browsing, evaluation, and
use of ontologies. These features include capabilities for which there
is no specific correlate in current technology for ontology
management. These features include online community-based review and
annotation of ontologies, support for declaring mappings between
ontologies, and the push of information about ontology changes
directly to a user's desktop. These novel capabilities for ontology
access and use have the potential to address pressing end-user needs
and to engage entire research communities in the development,
evolution, and evaluation of controlled terminologies and
ontologies. Our work offers possibilities for the creation of a new
kind of collaborative, community-based approach to the dissemination
and use of ontologies. The key question is whether our user community
is ready for such capabilities. The issue is one of sociology and
culture as much as it is one of technology. This talk will describe
the BioPortal for access to biomedical ontologies online, and will
outline our future plans for this novel, Web-based resource.
The online Ontolog events are open and free of charge. Anyone who is
interested, or (better still) who may have something to contribute, is
welcome. Refer to individual event details on their respective session
pages. The links are given, where you will find session agenda,
presentation abstract, conference call dial-in and other pertinent
details . This event's wiki page is at:
<http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2007_12_06>
*RSVP* by by emailing peter.yim...cim3.com so that they can prepare
enough resources to support everyone. (Kindly include your affiliation
and job title if you aren't already a member of the Ontolog
community.) Before participating, please also make sure you are aware
of the IPR policy (ref:
<http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid32>).
____________
SSP10: SYMBOLIC SYSTEMS FORUM
on Thursday, 6 December 2007, 4:15pm
Bldg. 380:380C (Math Corner)
http://symsys.stanford.edu/ssp_events
"Learning to Recognize Facial Emotions:
Psychologists vs. Artists"
David C. Wilkins
Symbolic Systems and CSLI
http://www.stanford.edu/~dwilkins/
Psychologists and Artists have adopted very different approaches to
the cognitive task of learning to recognize facial emotions.
Psychologists teach mainly by showing classified example of of faces,
e.g., <http://www.PaulEkman.com>. Artists teach by the immersive
experience of drawing live models, e.g.,
<http://www.DrawTheFeeling.org/>. Which is better? This talk
presents these different approaches and describes our efforts to
identify metrics and design experiments to quantify the differences. A
Symbolic Systems seminar course on this topic is offered next quarter,
and is described at
<http://www.stanford.edu/~dwilkins/symbsys210.pdf>.
____________
INFORMATION SYSTEMS SEMINAR
on Thursday, 6 December 2007, 4:15pm-5:15pm
Packard 101
http://isl.stanford.edu/colloquium.html
"Exchanging Limits: Why Iterative Decoding Works"
Rüdiger Urbanke
School of Computer & Communication Sciences, EPFL
Most new communication standards employ sparse graph codes and
iterative decoding algorithms.
The standard way of designing such systems is to assume that the
codelength is infinite and that a fixed number of iterations is
performed. We then imagine that we perform more and more iterations
and consider the resulting performance limit. This is called the
``density evolution'' limit. But what happens if we exchange the two
limits and we first let the number of iterations tend to infinity and
then only consider increasing blocklengths? From simulations it seems
that these two limits are the same but what can we say
mathemathically? Proving that the exchange of limits is valid is not
only intellectually pleasing but is also a first step in considering
scaling laws for sparse graph codes.
I will consider the noise regime below the threshold computed by
density evolution. For sparse graph codes with large variable degrees,
expansion arguments provide an easy way to show that the exchange of
limits is valid.
The situation is more interesting for sparse graph codes with small
left degrees. In this case expansion arguments are no longer
sufficient and we have to rely on combinatorial arguments and specific
properties of the decoder.
This is joint work with Satish Korada, EPFL.
About the Speaker: Rüdiger L. Urbanke received the Diplomingenieur
degree from the Vienna Institute of Technology, Vienna, Austria, in
1990 and the M.S. and Ph.D. degrees in electrical engineering from
Washington University, St. Louis, MO,in 1992 and 1995
respectively. From 1995 to 1999, he held a position at the Mathematics
of Communications Department at Bell Labs. Since November 1999, he has
been a faculty member at the School of Computer & Communication
Sciences of EPFL, Lausanne, Switzerland. Dr. Urbanke is a recipient of
a Fulbright Scholarship. From 2000-2004 he was an Associate Editor of
the IEEE Transactions on Information Theory and he is currently on the
board of the series "Foundations and Trends in Communications and
Information Theory." He is a co-recipient of the IEEE Information
Theory Society 2002 Best Paper Award and the co-author (jointly with
Tom Richardson) on an upcoming book entitled "Modern Coding Theory."
____________
END MATERIAL
The CSLI Calendar appears weekly on most Wednesdays throughout the
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). To unsubscribe use the word unsubscribe instead of
subscribe. Problems with subscribing or unsubscribing should be sent
to owner-csli-calendar@csli.stanford.edu
The full current issue is at
http://cslicalendar.stanford.edu/current.shtml
and the archives at
http://cslicalendar.stanford.edu/Archive/
People on many of the CSLI computers can type 'help csli-calendar' to
see the current issue.
The CSLI Calendar is also posted each week to the su.events usenet
newsgroup (only available from computers on the Stanford network)
Information about CSLI's research program is available at
http://www-csli.stanford.edu/
For maps to the Stanford University rooms see
http://cslicalendar.stanford.edu/locations.shtml