[Date Prev][Date Next][Date Index]
CSLI Calendar, Wednesday, 6 February 2008, vol. 23:20
CSLI CALENDAR OF PUBLIC EVENTS
______________________________________________________________________
6 FEBRUARY 2008 Stanford Vol. 23, No. 20
______________________________________________________________________
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 6 FEBRUARY 2008 TO 15 FEBRUARY 2008
WEDNESDAY, 6 FEBRUARY 2008
3:30pm SRI CCB Seminar Series [6-Feb-08]
AE201, SRI International
"Little b: A Symbolic Language For Building Modular Models of Biology"
Aneil Mallavarapu
Systems Biology, Harvard Medical School
Abstract below
4:00pm UC Berkeley School of Information Lecture [6-Feb-08]
110 South Hall (Berkeley)
"Combating the Participation Gap: Why New Media Literacy Matters"
Henry Jenkins
http://www.ischool.berkeley.edu/about/events/dls20080206
Abstract below
4:15pm EE380: Computer Systems Lab Colloquium [6-Feb-08]
Gates B01
"Program representation using a domain workbench"
Charles Simonyi
Intentional Software Corporation
http://ee380.stanford.edu/contents.html
Abstract below
4:30pm MS&E 472: Entrepreneurial Thought Leader Seminar [6-Feb-08]
Skilling Auditorium
Christine Benninger
President, Humane Society Silicon Valley Mixer
http://www.stanford.edu/class/msande472/
5:15pm Linguistics Department Colloquium
Margaret Jacks Hall 460:126
"Probabilistic Knowledge in Human Language Comprehension and
Production"
Roger Levy
UCSD
http://www-linguistics.stanford.edu/colloq/
Abstract below
6:00pm MIT Club Talk [6-Feb-08]
Berkeley City Club (Berkeley)
"On Organic Interfaces"
Victor Zue
Computer Science and Artificial Intelligence Lab, MIT
http://www.csail.mit.edu/
http://www.mitcnc.org/Events_Single.asp?eventID=3D1375
($20 for MITCNC members, $30 for non-members)
THURSDAY, 7 FEBRUARY 2008
12 noon CSLI CogLunch [7-Feb-08]
Cordura Hall 100
"Metaphysics, Time, and Rationality"
Jean-Pierre Dupuy
Stanford University
http://www-csli.stanford.edu/events/Coglunch/
Abstract below
4:00pm CCRMA Hearing Seminar [7-Feb-08]
CCRMA Seminar Room, The Knoll
"On Making Machines that Listen"
Paris Smaragdis
Adobe Labs
http://ccrma-mail.stanford.edu/mailman/listinfo/hearing-seminar
Abstract below
4:00pm PARC Forum [7-Feb-08]
George Pake Auditorium at PARC
"Kiva.org: a transparent, web-based microfinance system"
Premal Shah
Kiva.org
http://www.parc.com/forum/
4:00pm UC Berkeley CIS Seminar [7-Feb-08]
Soda Hall 405 (UC Berkeley)
"V-fold penalization: an alternative to V-fold cross-validation"
Sylvain Arlot
Universite Paris-Sud
http://www.cs.berkeley.edu/~rakhlin/cis-seminar
Abstract below
4:15pm SSP10: Symbolic Systems Forum [7-Feb-08]
Bldg. 380:380C (Math Corner)
"Meta-Metaphysics Or: On What There Isn't"
Alexis Burgess
Philosophy Department,
http://symsys.stanford.edu/ssp_events
Abstract below
4:15pm Information Systems Seminar [7-Feb-08]
Packard 101
"Counter braids: An efficient counter architecture for
measuring network traffic"
Balaji Prabhakar
Stanford University
http://isl.stanford.edu/colloquium.html
Abstract below
4:15pm Neurobiology of Disease Seminar Series [7-Feb-08]
Munzer Auditorium, Beckman B060
"The Cruelest Cut of All: Caspase Cleavage of Htt and the
Pathogenesis of HD"
Michael Hayden
http://nis-seminars.stanford.edu/
5:30pm Stanford Semantics and Pragmatics Workshop [7-Feb-08]
Margaret Jacks Hall 460:126
"Real-Time Evidence for Semantic Coercion"
Brian McElree
NYU
(cosponsored with SPLaT!; refreshments at 5:15pm)
http://www-linguistics.stanford.edu/semgroup/
Abstract below
FRIDAY, 8 FEBRUARY 2008
12 noon Ethics@Noon [8-Feb-08]
Bldg. 100:101K
Stephen Schneider
Woods Institute / Biological Sciences
http://stephenschneider.stanford.edu/
http://www.stanford.edu/dept/EIS/lectures_ethics.html
12:30pm CS547: Human-Computer Interaction Seminar [8-Feb-08]
Gates B01
"User Interaction With a Personalized Mobile Recommender of
Leisure Activities"
Kurt Partridge and Ellen Isaacs
PARC
http://hci.stanford.edu/cs547/
Abstract below
2:00pm GRAI Seminar [8-Feb-08]
Gates 120
"Mobile Visual I/O"
Kari Pulli
Nokia Research
http://cs.stanford.edu/people/theobalt/GRAI.html
Abstract below
2:00pm Berkeley International Computer Science Institute [8-Feb-08]
ICSI, Rm 5A (UC Berkeley)
"Speaker Role Detection in Meetings using Social Network"
Neha Priyadarshini Garg
Abstract below
2:45pm Berkeley International Computer Science Institute [8-Feb-08]
ICSI, Rm 5A (UC Berkeley)
"Cross-lingual information distillation using statistical
classification methods"
Adish Kumar Singla
http://www.icsi.berkeley.edu/
Abstract below
3:00pm Berkeley Information Access Seminar [8-Feb-08]
107 South Hall (Berkeley)
"An Open Context for Archeology."
Eric Kansa
UC Berkeley
http://courses.ischool.berkeley.edu/i296a-1/s08/schedule.html
Abstract below
3:15pm Friday Cognitive Seminar [8-Feb-08]
Jordan Hall 420:050
Title to be announced
Greg Corrado
Neuroscience, Stanford
http://www-psych.stanford.edu/events_cognitive.html
4:15pm CS545: Stanford-HP InfoSeminar [8-Feb-08]
Gates B03
"Keyword Search on Graph-Structured Data"
S. Sudarshan
IIT
http://infolab.stanford.edu/infoseminar/
Abstract below
MONDAY, 11 FEBRUARY 2008
11:00am Media X Seminar: Workgroup Protocols and Algorithms [11-Feb-08]
Wallenberg Hall Learning theater (Bldg. 160)
"Social Messes: Using Visual Language To Help With Wicked Problems"
Robert Horn
Visiting Scholar, CSLI
http://www.stanford.edu/~rhorn/
http://mediax.stanford.edu/events_calendar/calendar.html
12 noon Work, Technology and Organization Colloquium [11-Feb-08]
Terman 217
Title to be announced
Beth Bechky
Management, UC Davis
(Joint WTO-SCANCOR Seminar)
http://www.gsm.ucdavis.edu/Faculty/index.aspx?id=409
http://www.stanford.edu/group/WTO/
12:45pm CIS/SLATA [11-Feb-08]
Law School 280B
"Virtual Design and Trustworthy Signals"
Judith Donath
MIT Media Lab
http://cyberlaw.stanford.edu/
Abstract below
4:00pm UC Berkeley Linguistics Colloquium [11-Feb-08]
182 Dwinelle Hall (Berkeley)
"Two to tangle: The origins and structure of Gurindji Kriol,
an Australian mixed language"
Felicity Meakins
U. Manchester
http://linguistics.berkeley.edu/events/
4:00pm UC Berkeley Working Group in the Philosophy of Mind [11-Feb-08]
Tolman 5101 (Berkeley)
"Reading and discussion of: Jerry Fodor, 'The Mind Doesn't
Work That Way', chaps. 4 and 5"
Reading group
http://neurophilosophy.berkeley.edu/meetings.html
TUESDAY, 12 FEBRUARY 2008
4:00pm PACS Seminar [12-Feb-08]
Haas Center for Public Service
"The Internet and Giving: Which Experiments Matter in 2008?"
Lucy Bernholz
Blueprint Research & Design
(limited space so rsvp)
http://pacscenter.stanford.edu
Abstract below
4:15pm Mathematical Logic Seminar [12-Feb-08]
Bldg. 60:119
"The Essential Proposition: Frege on Identity Statements"
Robert May
UC Davis
http://www-logic.stanford.edu/
Abstract below
4:15pm ENGR110/210: Perspectives in Assistive Technology [12-Feb-08]
Meyer Library 124
"Patentability and Infringement - Debunking the Common Myth"
Jeffrey M. Schox
Registered Patent Attorney and Stanford Lecturer
http://www.stanford.edu/class/engr110/
WEDNESDAY, 13 FEBRUARY 2008
4:15pm EE380: Computer Systems Lab Colloquium [13-Feb-08]
Gates B01
"Interaction Techniques Using the Wii Remote"
Johnny Chung Lee
CS Department, Carnegie Mellon University
http://ee380.stanford.edu/contents.html
Abstract below
4:30pm MS&E 472: Entrepreneurial Thought Leader Seminar [13-Feb-08]
Skilling Auditorium
Brett Crosby
Group PMM, Google Analytics
http://www.stanford.edu/class/msande472/
6:30pm SF Bay ACM Data Mining SIG [13-Feb-08]
SAP LABS, Building D, 3410 Hillview Avenue, Palo Alto, CA
"SILVIA:
A Practical Hybrid Approach for Conversational Intelligence"
Leslie Spring
CTO and Founder, Cognitive Code Corporation
http://sfbayacm.org/dmsig.php
Abstract below
THURSDAY, 14 FEBRUARY 2008
12 noon CIS/SLATA [14-Feb-08]
Tresidder Union, Cypress Room N-S
"The Ethics of Social Networking"
Panel
http://cyberlaw.stanford.edu/
Abstract below
12:15pm Stanford Networking Seminar [14-Feb-08]
Packard 101
Title to be announced
Jon Peterson
NeuStar
http://netseminar.stanford.edu/
3:15pm Philosophy Department Colloquium [14-Feb-08]
Bldg. 90:92Q
"Aristotle on Responsibility for One's Character"
Pierre Destrée
Université De Louvain
Co-sponsor: Classics
http://www-philosophy.stanford.edu/ce.html
4:15pm SSP10: Symbolic Systems Forum [14-Feb-08]
Bldg. 380:380C (Math Corner)
"New Tools, New Rules: Protocols, Algorithms,and the Future of Work"
Stanley Rosenschein
CSLI
http://www.quindi.com/about.htm#s
http://symsys.stanford.edu/ssp_events
Abstract below
4:15pm Information Systems Seminar [14-Feb-08]
Packard 101
"Stability and asymptotic optimality of generalized MaxWeight policies"
Sean Meyn
University of Illinois
http://isl.stanford.edu/colloquium.html
Abstract below
4:15pm Neurobiology of Disease Seminar Series [14-Feb-08]
Munzer Auditorium, Beckman B060
"Massive and Specific Dysregulation of Hippocampal Circuitry
in Temporal Lobe Epilepsy"
Douglas Coulter
http://www.med.upenn.edu/ins/faculty/coulter.htm
http://www-med.stanford.edu/sbrc/calendar/
FRIDAY, 15 FEBRUARY 2008
12 noon Speech Lunch [15-Feb-08]
Phonetics Lab
"Hidden Conditional Random Fields for Disfluency Detection"
Yun-Hsuan Sung
12:30pm CS547: Human-Computer Interaction Seminar [15-Feb-08]
Gates B01
"Visualizations and Ambient Interfaces"
S. Joy Mountford
Yahoo!
http://www.idbias.com/people.html
http://hci.stanford.edu/cs547/
Abstract below
2:00pm GRAI Seminar [15-Feb-08]
Gates 104
Title to be announced
James Davis
UC Santa Cruz
http://cs.stanford.edu/people/theobalt/GRAI.html
3:15pm Friday Cognitive Seminar [15-Feb-08]
Jordan Hall 420:050
Title to be announced
Allison Master
Psychology, Stanford
http://www-psych.stanford.edu/events_cognitive.html
4:15pm CS545: Stanford-HP InfoSeminar [15-Feb-08]
Gates B03
"Towards a Distributed Web Search Engine"
Ricardo Baeza-Yates
Yahoo! Spain and Chile
http://www.dcc.uchile.cl/~rbaeza/
http://infolab.stanford.edu/infoseminar/
Abstract below
____________
Stanford Blood Center: Shortage of O+, O-, A+, 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.
____________
ANNOUNCEMENT
The UC Berkeley Townsend Center Working Group in Philosophy of the
Mind is starting another semester of activity. In this semester, they
will be continuing their readings on their chosen topic for this year:
"Evolution and Mind: Optimality Design and Function". In addition,
they have several guest speakers, from Berkeley and from other
institutions. For more information see
http://neurophilosophy.berkeley.edu/meetings.html
____________
UPCOMING
The sixth annual Media X meeting will be on 3-4 March 2008. See
http://mediax.stnaford.edu/ for more information.
The ninth annual Semantics Fest will be on 14 March 2008. See
http://www-linguistics.stanford.edu/semgroup/eventschedule.html
____________
SRI CCB SEMINAR SERIES
on Wednesday, 6 February 2008, 3:30pm
AE201, SRI International
"Little b: A Symbolic Language For Building Modular Models of Biology"
Aneil Mallavarapu
Systems Biology, Harvard Medical School
The little b project is an effort to provide an open source language
which allows scientists to build mathematical models of complex
systems. Our goal is to stimulate widespread sharing and reuse of
biological models.
Today modellers retread a path walked by software and electronics
engineers: the problem of how to specify complex systems concisely,
comprehensibly and reusably. Engineers develop large complex systems
by marshalling time-tested tools and approaches which subdue
complexity through modular interfaces and hierarchical design. For
biological models, a new set of problems exists: how to tame
combinatorial complexity arising from molecular interactions, how to
deal with uncertainty and changing knowledge, and how to permit
multiple analyses of the same formal representation of biological
knowledge.
Little b is a language designed to enable terse, readable and modular
descriptions of cellular and molecular systems. The language extends
Common Lisp with a reasoner, symbolic mathematics and pattern-matching
algorithms. Lisp's powerful metaprogramming capabilities enable a
"literate modelling" style in which the model serves as a formal
specification of biological knowledge. I'll discuss the little b
approach, and supporting tools, with examples which demonstrate how
the language handles multicellular and molecular complexity.
About the Speaker: Dr. Mallavarapu directs the little b project, an
open source effort aimed at developing tools for modular, shareable
mathematical models of biology. Prior to joining Harvard, he worked at
Millennium Pharmaceuticals, where he developed several software and
wetware technologies used in the company's genomics-based drug
discovery platform. He initiated a pathway-centric approach to
genomic information management. Dr. Mallavarapu's training is in cell
biology and biochemistry. He developed a number of optical
technologies for marking, perturbing and visualizing protein dynamics
in living cells during PhD work with Tim Mitchison at UCSF, and
earlier predoctoral work with Dan Jay at Harvard. He is currently
Senior Research Scientist at the Department of Systems Biology at
Harvard Medical School.
____________
BERKELEY SCHOOL OF INFORMATION DISTINGUISHED LECTURE
on Wednesday, 6 February 2008, 4:00pm - 5:30pm
101 South Hall (Berkeley)
http://www.ischool.berkeley.edu/about/events
"Combating the Participation Gap: Why New Media Literacy Matters"
Henry Jenkins
According to recent studies by the Pew Center on the Internet And
American Life, more than half of American teens online have produced
media content and about a third have circulated media that they have
produced beyond their immediate friends and family. These statistics
reflect the growing importance of participatory culture in the
everyday lives of American young people. Work across a range of
disciplines suggest that these emerging forms of participatory culture
are important sites for informal learning and may be the crucible out
of which new conceptions of civic engagement are emerging. Drawing on
insights from a recent white paper produced for the MacArthur
Foundation, this talk will discuss the need to develop new forms of
media literacy pedagogy which reflects this context of a participatory
culture, materials which both respond to the ethical challenges
confronted by those teens who are already producing and circulating
their own media as well as the challenges confronting those youth who
are excluded from participation in these on-line worlds as a
consequence of lack of access to technologies, skills, competencies,
and cultural experiences taken for granted by their
contemporaries. These issues can not be understood through a simple
opposition between digital natives and digital immigrants, but rather
require us to dig deeper into the diverse range of experiences young
people have online and the range of different interactions between
adults and teens in these new participatory culture. In the course of
the presentation, I will be sharing a range of curricular materials
and activities being developed by MIT's Project nml to support the
teaching of these new social skills and cultural competencies.
About the Speaker: Henry Jenkins is the Co-Director of the MIT
Comparative Media Studies Program and the Peter de Flores Professor of
Humanities. He is also the author and/or editor of twelve books on
various aspects of media and popular culture, including Convergence
Culture: Where Old and New Media Collide, Fans, Bloggers and Gamers:
Exploring Participatory Culture, The Wow Climax: Tracing the Emotional
Impact of Popular Culture, Textual Poachers: Television Fans and
Participatory Culture, Hop on Pop: The Politics and Pleasures of
Popular Culture, and From Barbie to Mortal Kombat: Gender and Computer
Games. Jenkins writes regularly about media and cultural change at his
blog, henryjenkins.org. He is one of the principal investigators for
The Education Arcade, a consortium of educators and business leaders
working to promote the educational use of computer and video games and
of the Knight Center for Future Civic Media, a joint effort with the
MIT Media Lab to use new media to enhance how people live in local
communities. He is one of the principle investigators for GAMBIT, a
lab focused on promoting experimentation through game design, and of
Project nml, a MacArthur Foundation funded project that develops
curricular materials focused on promoting the social skills and
cultural competencies needed to become a full participant in the new
media era. Jenkins has a MA in Communication Studies from the
University of Iowa and a PhD in Communication Arts from the University
of Wisconsin-Madison.
____________
EE380: COMPUTER SYSTEMS LAB COLLOQUIUM
on Wednesday, 6 February 2008, 4:15pm to 5:30pm
Gates B01
http://ee380.stanford.edu/contents.html
"Program representation using a domain workbench"
Charles Simonyi
Intentional Software Corporation
A Domain Workbench is a new category of tool that can be used to store
and edit program-like structures that can be projected in arbitrary
notations. The talk will discuss Intentional Programming as the
methodology motivating the requirements. We will look at the issues
from the theoretical point of view of how to represent great
complexity efficiently, from the management point of how to involve
domain experts more directly into the software creation process, and
from the software engineering point of view of how to specify and use
a tool that can bridge the needs of domain experts, programmers, and
program generating software.
____________
LINGUISTICS DEPARTMENT COLLOQUIUM
on Wednesday, 6 February 2008, 5:15pm
Margaret Jacks Hall 460:126
http://www-linguistics.stanford.edu/colloq/
"Probabilistic Knowledge in
Human Language Comprehension and Production"
Roger Levy
UCSD
The talk covers two fundamental issues, one each in language
comprehension and production: what determines the difficulty of
comprehending a given word in a given sentence, and what factors
influence the choice that a speaker makes when it is possible to
express a meaning more than one way? The first half of the talk
presents the surprisal theory of processing difficulty, building on
Hale (2001), based on the premise that sentence comprehension involves
the rational and fully incremental application of probabilistic
knowledge. On this theory, the comprehender's probabilistic
grammatical knowledge determines expectations about the continuations
of a sentence at multiple structural levels; and these expectations
determine the difficulty of processing the words that are actually
encountered. I show how this theory can be applied to a number of
results in online sentence comprehension (Konieczny, 2000; Konieczny &
Doering, 2003; Jaeger et al., 2005) that are problematic for
memory-oriented theories such as the Dependency Locality Theory
(Gibson 1998, 2000) or Similarity-Based Interference (Gordon et al.,
2001, 2004; Lewis et al., 2006), yet were not covered by previous
probabilistic theories because the results do not involve resolution
of structural ambiguity. Next, I describe how surprisal can be
derived in multiple ways from optimality principles, and present
results supporting the claim that processing times in language
comprehension truly are linear in negative log-probability. I also
describe recent experimental results on the processing of extraposed
relative clauses showing that a number of results reported by Gibson &
Breen (2005) originally interpreted in terms of locality and phrasal
adjacency can be subsumed and generalized under the rubric of
surprisal.
The idea that probabilistic expectations drive processing difficulty
leads to the final proposal of the talk: that speakers make choices in
language production such that their utterances tend toward an optimal,
uniform level of information density. This last part of the talk
introduces the basic theory of uniform information density, and
presents an empirical study and model using the parsed Switchboard
corpus to investigate speaker choice in optional relativizer omission,
such as (1) below:
How big is the family (that) you cook for __?
We find that speakers tend to use the optional relativizer "that" more
often when the information density of the onset of the relative clause
is higher. These results provide evidence in support of uniform
information density as a locus of optimal production decisions.
____________
CSLI COGLUNCH
on Thursday, 7 February 2008, 12 noon - 1:00pm
Cordura Hall 100
http://www-csli.stanford.edu/events/Coglunch/
"Metaphysics, Time, and Rationality"
Jean-Pierre Dupuy
Stanford University
One of the oldest problems in Metaphysics is Diodorus' Master
Argument. Diodorus derived a contradiction from the conjoint assertion
of three propositions: 1) Every past truth is necessary; 2) The
impossible does not follow from the possible; 3) Something is possible
which neither is nor will be true. Diodorus purported to derive the
negation of 3), as if he wanted "to infect the future with the
necessity which belongs to the past."
Following the lead of Jules Vuillemin, Necessity or Contingency. The
Master Argument, CSLI Publications, Stanford University, 1996, who
construed certain of the modalities as essentially tensed, and drawing
on Henri Bergson's metaphysics of temporality, I have been able to
defend free will against the Master argument in a novel way (see J.-P.
Dupuy, "Philosophical Foundations of a New Concept of Equilibrium in
the Social Sciences: Projected Equilibrium", Philosophical Studies,
100, 2000, p. 323-345.)
The talk will not delve into the technical details of the
demonstration but will rather explore the relevance and the
implications of the proposed solution for the theory of rationality.
We will show in particular how it solves elegantly some of the most
nagging paradoxes of Rational Choice Theory (Newcomb, Backward
Induction) and helps tackle such pressing issues as the efficiency and
ethics of nuclear deterrence or the foundations of the Precautionary
Principle.
About the Speaker: Jean-Pierre Dupuy is Professor of philosophy, Ecole
Polytechnique, Paris, and founding director of C.R.E.A. (Centre de
Recherche en Epistemologie Applique), the philosophical research group
of the Ecole Polytechnique, and Full Professor (1/3rd time),
Departments of French and, by courtesy, Political Science, Stanford
University. He is also a Stanford C.S.L.I. Researcher, and is
affiliated with the Stanford Science, Technology, and Society Program,
the Symbolic Systems Forum, the Anthropology Department, and the
Religious Studies Department. He is a member of the French Academy of
Technology.
____________
CCRMA HEARING SEMINAR
on Thursday, 7 February 2008, 4:00pm
CCRMA Seminar Room, The Knoll
http://ccrma-mail.stanford.edu/mailman/listinfo/hearing-seminar
I'm really pleased to announce that Paris Smaragdis will be discussing
his work to build machines that listen. Paris was at the MIT Media
Lab, and then Mitsubishi's lab in Boston, and now he is a researcher
at Adobe, working on all aspects of sound.
Paris is one of the first people to suggest that there are a range of
statistical models that one can use to decompose sound. People with a
statistical background think about the independent components (ICA) of
a sound, as measured by Bayesian theory. People like Al Bregman think
about the holistic aspects of sound, and how our experience causes us
to hear different parts of the sound field as single objects (ASA).
Over the last 10 years, Paris has been studying the connection between
these two outlooks. Is ICA a subfield of ASA? Can one predict all
the aspects of auditory scene analysis (ASA) based on statistical
properties of sound? Paris is in a unique position to illustrate the
connections. --Malcolm Slaney
"On Making Machines that Listen"
Paris Smaragdis
Adobe Labs
Enabling machines to perceive the world using various modalities them
is one of the holy grails of artificial intelligence. In this talk I
will present some of my research on creating machines that do as such
by listening. I will discuss some of the unique difficulties in this
field and present a thread of research which spans a range of
computational techniques relating to signal processing, machine
learning and cryptography. This research will be introduced in the
context of classic audio problems such as time/frequency analysis,
music transcription, source separation, recognition in mixtures and
more. I'll show how this work generalizes and finds applications to
other domains, such as computer vision, and what it takes to move it
from the whiteboard to the real-world.
About the Speaker: Paris Smaragdis is a senior research scientist at
Adobe Systems. He completed his graduate and postdoctoral studies at
MIT where he conducted his research in computational perception.
Prior to Adobe he was a research scientist at Mitsubishi Electric
Research Labs, during which time he was selected by the MIT Technology
Review as one of the top 35 young innovators of 2006. Paris' research
interests lie in the intersection of machine learning and signal
processing, especially as they apply to problems of computational
audition.
____________
UC BERKELEY CIS SEMINAR
on Thursday, 7 February 2008, 4:00pm-5:00pm
Soda Hall 405 (UC Berkeley)
http://www.cs.berkeley.edu/~rakhlin/cis-seminar
"V-fold penalization: an alternative to V-fold cross-validation"
Sylvain Arlot
Universite Paris-Sud
We study the efficiency of $V$-fold cross-validation (VFCV) for model
selection from the non-asymptotic viewpoint, and suggest an
improvement on it, which we call "$V$-fold penalization".
First, considering a particular (though simple) regression problem, we
prove that VFCV with a bounded $V$ is suboptimal for model selection.
The main reason for this is that VFCV "overpenalizes" all the more
that $V$ is large. Hence, asymptotic optimality requires $V$ to go to
infinity. However, when the signal-to-noise ratio is low, it appears
that overpenalizing is necessary, so that the optimal $V$ is not
always the larger one, despite of the variability issue. This is
confirmed by some simulated data.
In order to improve on the prediction performance of VFCV, we define a
new model selection procedure, called "$V$-fold penalization"
(penVF). It is a $V$-fold subsampling version of Efron's bootstrap
penalties, so that it has the same computational cost as VFCV, while
being more flexible. In a heteroscedastic regression framework,
assuming the models to have a particular structure, we prove that
penVF satisfies a non-asymptotic oracle inequality with a leading
constant close to 1. In particular, this implies adaptivity to the
smoothness of the regression function, even with a highly
heteroscedastic noise. Moreover, it is easy to overpenalize with
penVF, independently from the $V$ parameter. According to a simulation
study, this results in a significant improvement on VFCV in
non-asymptotic situations.
____________
SSP10: SYMBOLIC SYSTEMS FORUM
on Thursday, 7 February 2008, 4:15pm
Bldg. 380:380C (Math Corner)
http://symsys.stanford.edu/ssp_events
"Meta-Metaphysics Or: On What There Isn't"
Alexis Burgess
Philosophy Department, Stanford
Metaphysics is supposed to be the most general study of "what there
is". Thus debates in metaphysics often take the following form.
Professor X asserts that something exists (or some things exist).
Professor Y denies that that thing exists (or those things exist).
Argument ensues. Sometimes a consensus emerges among the philosophical
community about which side is right; more often, though, as one would
expect in philosophy, agreement proves elusive. Yet there is a
striking degree of agreement among contemporary analytic philosophers
about what such pairs of professors are saying; and in particular,
about what we mean by the word 'existence'. According to the orthodoxy
derived from the work of W. V. Quine, existence is just what's
expressed by the so-called existential quantifier of first-order
logic. It is a primitive, unanalyzable notion. Unfortunately, the
orthodoxy offers no compelling solution to a problem that has
frustrated philosophers since antiquity, and which, ironically, Quine
himself brought to the fore in his seminal essay 'On What There
Is'. The problem is that it is hard to see how Professor Y could ever
be right. For how can we deny the existence of something without
referring to that thing, thereby presupposing its existence in our
very denial? In this presentation, I'll outline some famous attempts
to answer this meta-metaphysical challenge to the coherence of
metaphysical debate, present the arguments against them, and then
begin to develop a novel way out of the problem: what might be called
Defeatism about intuitively true, negative existential statements.
____________
INFORMATION SYSTEMS SEMINAR
on Thursday, 7 February 2008, 4:15pm-5:15pm
Packard 101
http://isl.stanford.edu/colloquium.html
"Counter braids: An efficient counter architecture for
measuring network traffic"
Balaji Prabhakar
Stanford University
Measuring network flow sizes is important for accounting/billing,
network forensics and security. Obtaining per-flow measurements is
hard because it requires fast and large memories, which are
essentially unavailable. Therefore, current approaches aim to obtain
approximate flow counts; that is, to detect large "elephant" flows and
then measure their sizes.
We present a novel method for per-flow traffic measurement that is
fast, highly memory efficient and accurate. At the core of this method
is an architecture, which we call "counter braids," which is inspired
by sparse random graph codes. We show: (i) The optimality of counter
braids in that the space needed to losslessly store flow sizes equals
the entropy lower bound. Since network traffic has an entropy rate
less than 3 bits per flow, this makes possible an implementation in
on-chip SRAM. (ii) A low-complexity message-passing estimation
algorithm, that recovers flow sizes with vanishing error at link
speed. Evaluation on Internet traces demonstrates that almost all flow
sizes are estimated error-free with only 5 bits per flow.
The topic is thematically related to graphical models, sparse random
graph codes and compressed sensing.
Joint work with: Yi Lu, Andrea Montanari, Sarang Dharmapurikar and
Abdul Kabbani
About the Speaker: Balaji Prabhakar is an Associate Professor of
Electrical Engineering and Computer Science at Stanford University.
Balaji is interested in network algorithms, in scaleable methods for
network performance monitoring and simulation, in wireless (imaging)
sensor networks, stochastic network theory and information theory. He
has designed algorithms for switching, routing, bandwidth
partitioning, load balancing, and web caching.
Balaji has been a Terman Fellow at Stanford University and a Fellow of
the Alfred P. Sloan Foundation. He has received the CAREER award from
the National Science Foundation, the Erlang Prize from the INFORMS
Applied Probability Society, and the Rollo Davidson Prize awarded to
young scientists for their contributions to probability and its
applications.
____________
STANFORD SEMANTICS AND PRAGMATICS WORKSHOP
on Thursday, 7 February 2008, 5:30pm
Margaret Jacks Hall 460:126
http://www-linguistics.stanford.edu/semgroup/
"Real-Time Evidence for Semantic Coercion"
Brian McElree
NYU
(cosponsored with SPLaT!; refreshments at 5:15pm)
To what degree are the meanings of natural language expressions
compositional? Strongly compositional approaches to meaning maintain
that each operation of semantic composition corresponds to an
operation of syntactic composition, so that the interpretation of an
expression can be fully determined by the meanings of its constituents
and by the syntactic way the constituents are combined. Expressions
containing syntax-semantic mismatches challenge strong
compositionality, as they may require semantic operations that do not
correspond to syntactic operations.
The pairing of an aspectual verb, such as begin or finish, with a
non-event-denoting object (e.g., ...began the book or ...finished the
beer) has been argued to contain one type of syntax-semantic
mismatch. The interpretation such verb phrases has been hypothesized
to involve a process of complement coercion, a non-syntactic process
that converts the entity-denoting object into an event description
that satisfies the selectional demands of the verb. I will present
studies using a variety of real-time processing measures (self-paced
reading and eye-tracking measures, magnetoencephalographic [MEG]
patterns, and speed/accuracy trade-off measures) that (1) show that
expressions hypothesized to involve complement coercion are indeed
more taxing to process than expressions with transparent
syntax-semantic mappings (including expressions involving metonymies,
such as ...read Dickens) and (2) suggest that the observed processing
costs reflect the operations involved in building an event description
to resolve the semantic type mismatch between the verb and
object. Time permitting, I will discuss differences between complement
coercion and other types of syntax-semantic mismatches, which do not
engender the same type of processing cost.
____________
CS547: HUMAN-COMPUTER INTERACTION SEMINAR
on Friday, 8 February 2008, 12:30-2:00pm
Gates B01
http://hci.stanford.edu/cs547/
"User Interaction With a Personalized Mobile Recommender of
Leisure Activities"
Kurt Partridge and Ellen Isaacs
PARC
How do you decide where to go for fun? In this talk, we will discuss
a mobile system that helps people discover places to go during their
leisure time by recommending nearby venues such as restaurants, shops,
museums, events, and parks. This system, codenamed Magitti,
personalizes its recommendations by combining explicit general
preferences, collaborative filtering, and predictions of your current
activity and interests from past behavior. We will give an overview
of the initial fieldwork, system design, and user evaluation, and
present a demonstration of the system.
About the Speakers: Kurt Partridge is a researcher in the Ubiquitous
Computing Area in the Palo Alto Research Center (PARC). His research
spans a variety of areas, including context awareness, activity
modeling, location modeling, mobile device interaction, and wearable
computing. He is particularly interested in systems and devices that
blend naturally with people's everyday activities. Kurt received a
Ph.D. in Computer Science from the University of Washington in 2005.
Ellen Isaacs is an interaction designer and researcher in the
Socio-Technical and Interaction Research group at Palo Alto Research
Center. During her career, Ellen has moved between design, evaluation,
research, and management at a variety of Silicon Valley companies,
including Sun Microsystems, Electric Communities, Excite, AT&T Labs,
and most recently, Izix Consulting, her own interaction design
consulting firm. Her research interests have focused primarily on
systems that support collaboration and communication through a variety
of media. She is the co-author with Alan Walendowski of "Designing
From Both Sides of the Screen," a book about the process of
interaction design and the collaboration between interaction designers
and software engineers. Ellen is also an accomplished photographer who
sells her images professionally. She received her PhD in
psycholinguistics from Stanford University.
____________
GRAI SEMINAR
on Friday, 8 February 2008, 2:00pm
Gates 120
http://cs.stanford.edu/people/theobalt/GRAI.html
"Mobile Visual I/O"
Kari Pulli
Nokia Research
Mobile devices are the new frontier in computation, internet
connectivity, imaging, and computer graphics. For many people it
provides their first computer of any kind. For example in China there
are about 100 million PCs but over 400 million mobile phones. And the
sales of camera phones far surpass the number of digital or any other
types of consumer cameras. The visual capabilities of these devices
have progressed at amazing pace, and today the high-end smart phones
have large color screens, fast CPUs, even dedicated graphics HW
acceleration. We will cover the short history of mobile 3D graphics,
and introduce recent mobile graphics standards such as OpenGL ES, M3G,
and OpenVG; these provide visual output. In recent years photography
and graphics have begun to move closer to each other, we will also
discuss computational and contextual photography and augmented reality
as fertile research topics and present some results; these provide
visual input.
About the Speaker: Kari Pulli is a Research Fellow at Nokia Research
Center in Palo Alto, where he heads a team working on how to make good
use of cameras on mobile phones. He has been an active contributor to
several mobile graphics standards.
Kari has a PhD in computer science from University of Washington and
Lic. Tech., M.Sc., and MBA from University of Oulu, where he also
taught computer graphics as an adjunct faculty.
Before Nokia Kari has worked on graphics at Stanford University,
Alias|Wavefront, SGI, and Microsoft; during 2004-06 Kari was also a
Visiting Scientist at MIT.
____________
BERKELEY INTERNATIONAL COMPUTER SCIENCE INSTITUTE
on Friday, 8 February 2008, 2:00pm
Conference Room 5A, ICSI, 1947 Center Street, Berkeley
http://www.icsi.berkeley.edu/
"Speaker Role Detection in Meetings using Social Network"
Neha Priyadarshini Garg
Meetings are part of everyone's day to day life. Usually, minutes are
taken to record what took place during the meeting. However, this can
be avoided by recording the meetings and then later navigating through
the content of meeting. Previous work has shown that, due to the
serial nature of speech, navigation of such data is difficult, and
structural information is useful for easier navigation. Users can
benefit from higher level structural information, such as who lead the
discussion? Who implemented the decisions etc. Identification of
speaker roles can be helpful in providing such high level information
about meetings. Previous work on speaker role detection either mainly
focused on using interaction patterns or lexical information for
broadcast news or conversations.
In this work, we compare various features that can be used for
detecting speaker roles in the project-oriented and naturally
occurring meetings (such as in the AMI and ICSI meetings corpora). In
particular, we analyze how much the knowledge of social networks
present in the meeting is helpful to detect the speaker roles along
with textual information. Our results show that social network
analysis measures and lexical features are both useful for speaker
role detection in meetings and their combination results in the best
performance.
____________
BERKELEY INTERNATIONAL COMPUTER SCIENCE INSTITUTE
on Friday, 8 February 2008, 2:45pm
Conference Room 5A, ICSI, 1947 Center Street, Berkeley
http://www.icsi.berkeley.edu/
"Cross-lingual information distillation using statistical
classification methods"
Adish Kumar Singla
As the amount of information available to users increase, accessing
information in an efficient way becomes difficult. The goal of
information distillation is to extract useful pieces of information
for a user's query from massive multi-lingual audio and textual
document sources. In the framework of the DARPA GALE project, these
sources are English, Arabic and Mandarin newswire, blogs, broadcast
news and conversations. One approach for answering these queries from
non-English sources is mainly training models using annotated data
from English documents and using machine translation (MT) output for
the non-English sources while searching answers to queries. While, the
noise introduced by automatic translation reduces the accuracy of
distillation process, there is more annotated data available from the
English sources.
In this work, we first experimented with the word-based statistical
approaches for answering user's queries from Arabic and Mandarin
sources. To benefit from the data from the English side, and the
absence of automatic translation noise on the source language side, we
combined models from the source languages and English. We have shown
that, this approach resulted in 5%-15% absolute improvement on the
GALE Year-1 and Year-2 data sets over the previous work.
____________
BERKELEY INFORMATION ACCESS SEMINAR
on Friday, 8 February 2008, 3:00pm - 5:00pm
107 South Hall (Berkeley)
http://courses.ischool.berkeley.edu/i296a-1/s08/schedule.html
"An Open Context for Archaeology."
Eric Kansa
UC Berkeley
The common use by archaeologists of ubiquitous technologies such as
computers and digital cameras means that archaeological research
projects now produce huge amounts of diverse, digital documentation.
However, while the technology is available to collect this
documentation, we still largely lack community accepted dissemination
channels appropriate for such torrents of data. Open Context
(<http://www.opencontext.org>) aims to help fill this gap by providing
open access data publication services for archaeology. Open Context
has a flexible and generalized technical architecture that can
accommodate most archaeological datasets, despite the lack of common
recording systems or other documentation standards. Open Context
includes a variety of tools to make data dissemination easier and more
worthwhile. Authorship is clearly identified through citation tools,
a web-based publication systems enables individuals upload their own
data for review, and collaboration is facilitated through easy
download and other features. While we have demonstrated a potentially
valuable approach for data sharing, we face significant challenges in
scaling Open Context up for serving large quantities of data from
multiple projects.
This talk will explore future work with commercial service providers,
including Metaweb to expand these efforts with a much more robust data
sharing infrastructure.
About the Speaker: Eric Kansa is the Executive Director of the
School's Information and Service Design Program
(<http://isd.ischool.berkeley.edu>). He has a background in
anthropology, archaeology, and in open access data sharing for the
field sciences. He is cofounder and former Executive Director of the
Alexandria Archive Institute, and led development of Open Context, an
online system for sharing collections and field research in
archaeology and natural history. This follows a position on the
faculty of Harvard University, where he served as Lecturer and
Undergraduate Tutor for the Department of Anthropology. He graduated
from the University of California, San Diego with a BA in Cultural
Anthropology and continued his education at Harvard University
beginning in 1995. There, he earned his doctorate in 2001 and has
focused research efforts on open dissemination strategies, information
architectures for the social sciences, and intellectual property
frameworks for online scholarship. Eric is currently Convener of the
Society for American Archaeology's Digital Data Interest Group.
____________
CS545: STANFORD-HP INFOSEMINAR
on Friday, 8 February 2008, 4:15pm - 5:30pm
Gates B03
http://infolab.stanford.edu/infoseminar/
"Keyword Search on Graph-Structured Data"
S. Sudarshan
IIT
A variety of types of data, such as relational, XML and HTML data can
be naturally represented using a graph model of data, with entities as
nodes and relationships as edges. Graph representations are also a
natural choice for representing information extracted from
unstructured data, and for representing information integrated from
heterogeneous sources of data.
Keyword search is a natural way to retrieve information from graph
representations of data, especially in the common situation where the
graphs do not have a well-defined schema. Unlike text or Web search,
there is no natural notion of a document, and information about a
single conceptual entity may be split across multiple nodes. Answers
to queries are therefore usually modeled as trees that connect nodes
matching the keywords, with each answer tree having an associated
score. The goal of keyword search then is to find answers with the
highest scores. A number of systems, including the BANKS system
developed at IIT Bombay, are based on such a model for answering
keyword queries.
In this talk we first outline background material on keyword querying
on graph data, including models for ranking of answer trees. We then
focus on search algorithms for finding top-ranked answers. We outline
key issues in finding top-ranked answers, and present algorithms that
address the problem in the context of in-memory graphs. We then
consider the problem of search on external memory graphs, and briefly
outline our on-going work on external memory graph search based on a
multi-granular graph representation.
About the Speaker: S. Sudarshan received the Ph.D. from the Univ. of
Wisconsin, Madison in 1992. He was a Member of the Technical Staff in
the database research group at AT&T Bell Laboratories, from 1992 to
1995, and he has been at the Indian Institute of Technology (IIT),
Bombay since 1995. He spent a year on sabbatical at Microsoft
Research, USA in 2004-05.
Sudarshan's research interests center on database systems, and his
current research projects include keyword querying on databases,
processing and optimization of complex queries, and database support
for securing and testing applications. Sudarshan is a co-author of
the widely used textbook, Database System Concepts, 5th Ed., by
Silberschatz, Korth and Sudarshan.
____________
CIS/SLATA
on Monday, 11 February 2008, 12:45pm-2:00pm
Law School 280B
http://cyberlaw.stanford.edu/
"Virtual Design and Trustworthy Signals"
Judith Donath
MIT Media Lab
Much of what we want to know about others is not directly perceivable
- are you a nice person? did you really like the cake I baked? would
you be a good employee, spouse, president? We rely instead on
signals, which are perceivable features or actions that indicate the
presence of those hidden qualities. Yet not all signals are
reliable. It is beneficial for the con-man to seem nice, for the guest
to seem to like the burnt cake, for the unsuitable suitor to seem as
attractive as possible. While these deceptions benefit the deceiver,
they may be quite costly for the recipient. What keeps signals honest
-- and why are some signals more reliable than others?
Signaling theory provides a framework for understanding these
dynamics. In this talk I will introduce signaling theory and show how
it can be used for the design and analysis of social technologies. It
is especially well suited for this domain, for in mediated
interactions there are few qualities that can be directly observed:
everything is signal.
About the Speaker: Judith Donath is an Associate Professor at the MIT
Media Lab, where she directs the Sociable Media research group. Her
work focuses on the social side of computing, synthesizing knowledge
from fields such as graphic design, urban studies and cognitive
science to build innovative interfaces for online communities and
virtual identities. She is known internationally for pioneering
research in social visualization, interface design, and computer
mediated interaction. She created several of the early social
applications for the web, including the first postcard service ("The
Electric Postcard"), the first interactive juried art show ("Portraits
in Cyberspace") and an early large-scale web event ("A Day in the Life
of Cyberspace"). Her work has been exhibited at the Institute for
Contemporary Art in Boston and in several New York galleries; she was
the director of "Id/Entity", a collaborative exhibit of installations
examining how science and technology' are transforming
portraiture. Her current research focuses on creating expressive
visualizations of social interactions and on building experimental
environments that mix real and virtual experiences. She has a book in
progress about how we signal identity in both mediated and
face-to-face interactions. Professor Donath received her doctoral and
master's degrees in Media Arts and Sciences from MIT, her bachelor's
degree in History from Yale University, and has worked professionally
as a designer and builder of educational software and experimental
media. She is currently a Faculty Fellow at the Berkman Center of
Harvard Law School.
____________
PACS SEMINAR
on Tuesday, 12 February 2008, 4:00pm
Haas Center for Public Service
http://pacscenter.stanford.edu
(limited space so rsvp)
"The Internet and Giving: Which Experiments Matter in 2008?"
Lucy Bernholz
Blueprint Research & Design
The Center on Philanthropy and Civil Society (PACS Center) cordially
invites you to its next seminar on Tuesday, February 12, 2008, from
4-6 pm at the DK Room in the Haas Center for Public Service.
The presentation, entitled "The Internet and Giving: Which Experiments
Matter in 2008?" will feature Lucy Bernholz, Founder and President of
Blueprint Research & Design, Inc. Blueprint is a strategy consulting
firm specializing in program research and design for philanthropic
foundations. Bernholz is a special fellow of the Synergos Institute
and a member of the International Network of Strategic Philanthropy.
She is also on the Advisory Boards of the Philanthropic Initiative for
Racial Equality, YouthGive, Social Venture Partners Bay Area, the
National Philanthropic Trust, and The Grantmaking School.
Additionally, she serves on the Board of Directors of CompuMentor and
is an Advisory Fellow on the Markey Steering Committee of Impact
Manager at the Center on Philanthropy and Public Policy at the
University of Southern California. She holds an MA and Ph.D. from
Stanford University.
Space is limited. Please RSVP to <nataliec .. stanford.edu> as soon
as possible, so that we can put you on the list and send you the
readings.
____________
MATHEMATICAL LOGIC SEMINAR
on Tuesday, 12 February 2008, 4:15pm
Bldg. 60:119
http://www-logic.stanford.edu/
"The Essential Proposition: Frege on Identity Statements"
Robert May
UC Davis
There is a certain standard myth about Frege's interest in identity
statements, centering on the notion that his interest was motivated by
a desire to justify a particular philosophical theory of meaning. On
this accounting, answering why "Hesperus is Phosphorus" and "Hesperus
is Hesperus" don't mean the same thing led Frege to posit a
substantive account of the meaning of expressions - the doctrine of
sense and reference - and it is this doctrine that constitutes Frege's
enduring contribution. There is of course more than a grain of truth
to this, but at heart it is misleading as to Frege's real interest in
identity statements; namely the essential role they play in logicism,
especially in establishing the fundamental tenet that numbers are
"self-subsistent" logical objects. For Frege, the puzzle of identity
statements called for solution just in order to maintain this
result. In this paper, I will explore how this plays out in the
development of Frege's views on identity statements, proposing that
the evolution from a view which countenances both metalinguistic
identity and mathematical equality, to one in which there is a single
notion of objectual identity, tracks a change in Frege's conception of
the relation of language and content, from language structuring
content to language representing content. In the course of the
presentation, I will explore the answers to the following questions,
among others: Why did Frege initially adopt a metalinguistic view?
What caused him to change his view to one in which identity statements
express objectual identity? What role do identity statements play in
the logicist program? What is the significance of the puzzle, and what
is its origin? And how does "On Sense and Reference" fit into Frege's
oeuvre?
____________
EE380: COMPUTER SYSTEMS LAB COLLOQUIUM
on Wednesday, 13 February 2008, 4:15pm to 5:30pm
Gates B01
http://ee380.stanford.edu/contents.html
"Interaction Techniques Using the Wii Remote"
Johnny Chung Lee
CS Department, Carnegie Mellon University
As of December 2007, Nintendo has sold over 20 million Wii consoles
worldwide. This significantly exceeds the number of tablet PCs used
today according to even the most generous estimates of tablet PC
sales. This makes the Nintendo Wii remote one of the most common input
devices in the world. It also happens to be one of the most
sophisticated containing a 3-axis accelerometer and high-resolution
high-speed infrared camera. This is an incredible opportunity to
explore interaction techniques enabled by the Wii remote and to
develop new applications that could be instantly accessible to
millions of individuals around the world. Though only just a few weeks
old, the work I will present has received nearly 5 million unique
views and generated over 250,000 software downloads. In this talk, I
will show you how you can participate in these projects as well as
generate your own.
About the Speaker: Johnny Chung Lee is a PhD graduate student in
Human-Computer Interaction at Carnegie Mellon University. His research
interests are in developing techniques that can significantly increase
the practicality and accessibility of technology to researchers,
developers, and end users. His previous work spans a wide range of
topics including projector-based augmented reality, brain-computer
interfaces, kinetic typography, haptics, multi-channel audio, tangible
interfaces, and filmmaking.
____________
SF BAY ACM DATA MINING SIG
on Wednesday, 13 February 2008, 6:30pm - 9:00pm
SAP LABS, Building D, 3410 Hillview Avenue, Palo Alto, CA
http://sfbayacm.org/dmsig.php
"SILVIA:
A Practical Hybrid Approach for Conversational Intelligence"
Leslie Spring
CTO and Founder, Cognitive Code Corporation
This talk describes a top-down, product development oriented approach
to architecting a conversationally intelligent software
system. Cognitive Code's SILVIA platform is used as an example of a
system that is representative of this approach. The SILVIA platform
employs a number of known methods within its implementation. The
SILVIA platform also incorporates a number of new, unique methods, as
well as new applications of known methods. However, independent of
the platform's algorithmic components, the overall SILVIA architecture
and implementation was developed with an eye toward rapid product
prototyping and development. As such a system, SILVIA derives many of
its elements from disparate development disciplines such as game
engines and content management systems. It is proposed that a more
general hybrid system not only creates an environment conducive to
rapid development and deployment across multiple markets, but also may
exhibit the desirable property of emergent behavior via the
interaction of its integrated components.
About the Speaker: Leslie is the inventor and architect of Cognitive
Code's SILVIA platform. As CTO and Founder, he is responsible for the
technology and product development arm of the company. Leslie also
works with the company's other executives and fellow board members to
determine the overall direction and specific implementations of
Cognitive Code's business strategies. Leslie has been a professional
software developer since the early 80s, and has worked on everything
from hand coding machine-language 3d matrix transformations to
managing large-scale development efforts and software groups. Leslie
has been employed by and has consulted to videogame and entertainment
companies such as Electronic Arts, Disney, and Sony Pictures. His
responsibilities as Principal Engineer, Technical Director, and
Software Architect for various projects have included heading teams to
develop commercial products, as well as direct, hands-on development
of a variety of graphics algorithms, gaming systems, and integrated
development tools. More recently, he was responsible for creating the
software architecture of Screenblast, Sony's on-line content creation
and sharing service, and developed all of Screenblast's content
management systems and associated user tools. In addition, while at
Sony, Leslie developed four large-scale patented software technologies
and shepherded these technologies through the entire patent
process. At four years of age, Leslie saw the film, 2001: a Space
Odyssey. It was at that moment that Leslie began his lifelong interest
in and passion for artificial intelligence. It is that passion
combined with decades of experience in developing practical,
product-oriented systems that has culminated in the development of the
SILVIA platform, and the founding of Cognitive Code as a business.
____________
CIS/SLATA
on Thursday, 14 February 2008, 12 noon
Tresidder Union, Cypress Room N-S
http://cyberlaw.stanford.edu/
"The Ethics of Social Networking"
Panel
Has "online privacy" become an oxymoron, or are we bound to develop
a "new privacy" concept that prioritizes nuanced control of personal
information? Is this new concept tenable? What are the benefits and
risks of the standardization of social networking? What impact does
social networking technology have on interpersonal virtues? Are we
witnessing the flattening of social landscape by online networks? And
can one be a college student without Facebook?
The Stanford Center on Ethics and the Stanford Center for Internet and
Society invite you to a discussion of a host of ethical and social
concerns generated by the evolving culture of social networking,
particularly by Internet users' habits and Web 2.0 sites' practices
and strategies just in time for Valentine's Day.
Moderated by Dean Eckles, a research scientist and designer at Nokia
Research Center.
Panelists include
Dr. BJ Fogg, Director of Research and Design at Stanford's Persuasive
Technology Lab, and lecturer at the Computer Science Dept;
Jia Shen, Co-Founder and CTO of RockYou;
Shannon Vallor, Professor of Philosophy at Santa Clara University.
____________
SSP10: SYMBOLIC SYSTEMS FORUM
on Thursday, 14 February 2008, 4:15pm
Bldg. 380:380C (Math Corner)
http://symsys.stanford.edu/ssp_events
"New Tools, New Rules: Protocols, Algorithms,and the Future of Work"
Stanley Rosenschein
CSLI
http://www.quindi.com/about.htm#s
Over the last two decades, digital networks have grown in power, reach
and availability, bringing impressive productivity gains to the
workplace. These gains, however, are limited by traditional work
practices, which, when combined with new tools, often lead to rapid
saturation of human capacity. This talk asks what kind of human
protocols and work practices might be better suited to dynamic work in
a digital age, and can these protocols be explicitly designed, in
analogy with computer protocols and algorithms?
____________
INFORMATION SYSTEMS SEMINAR
on Thursday, 14 February 2008, 4:15pm-5:15pm
Packard 101
http://isl.stanford.edu/colloquium.html
"Stability and asymptotic optimality of generalized MaxWeight policies"
Sean Meyn
University of Illinois
It is shown that stability of the celebrated MaxWeight or back
pressure policies is a consequence of the following interpretation:
either policy is myopic with respect to a surrogate value function of
a very special form, in which the "marginal disutility" at a buffer
vanishes for vanishingly small buffer population. This observation
motivates the h-MaxWeight policy}, defined for a wide class of
functions h. These policies share many of the attractive properties of
the MaxWeight policy:
* Arrival rate data is not required in the policy.
* Under a variety of general conditions, the policy is stabilizing
when h is a perturbation of a monotone linear function, a monotone
quadratic, or a monotone Lyapunov function for the fluid model.
* A perturbation of the relative value function for a workload
relaxation gives rise to a myopic policy that is approximately
average-cost optimal in heavy traffic, with logarithmic regret.
The first results are obtained for a general Markovian network model.
Asymptotic optimality is established for a general Markovian
scheduling model with a single bottleneck, and homogeneous servers.
About the Speaker: Sean P. Meyn received the B.A. degree in
Mathematics Summa Cum Laude from UCLA in 1982, and the PhD degree in
Electrical Engineering from McGill University in 1987 (with
Prof. P. Caines, McGill University). After a two year postdoctoral
fellowship at the Australian National University in Canberra, Dr.
Meyn and his family moved to the Midwest. He is now a Professor in the
Department of Electrical and Computer Engineering, and a Research
Professor in the Coordinated Science Laboratory at the University of
Illinois. He is also an IEEE fellow.
Dr. Meyn has served on the editorial boards of several journals in the
systems and control, and applied probability areas. He was a
University of Illinois Vice Chancellor's Teaching Scholar in 1994; is
coauthor with Richard Tweedie of the monograph Markov Chains and
Stochastic Stability, Springer-Verlag, London, 1993; and received
jointly with Tweedie the 1994 ORSA/TIMS Best Publication In Applied
Probability Award. His new book, Control Techniques for Complex
Networks is published by Cambridge University Press.
He has held visiting positions at universities all over the world,
including the Indian Institute of Science, Bangalore during 1997-1998
where he was a Fulbright Research Scholar. During his latest
sabbatical during the 2006-2007 academic year he was a visiting
professor at MIT and United Technologies Research Center (UTRC). His
research interests include stochastic processes, optimization, complex
networks, and information theory. Current funding is provided by NSF,
Motorola, DARPA, and UTRC.
____________
CS547: HUMAN-COMPUTER INTERACTION SEMINAR
on Friday, 15 February 2008, 12:30-2:00pm
Gates B01
http://hci.stanford.edu/cs547/
"Visualizations and Ambient Interfaces"
S. Joy Mountford
Yahoo!
http://www.idbias.com/people.html
Big companies with access to immense amounts of data can offer some
insights into how big the information world around us really is. Joy
will show a series of compelling visualizations that her Design
Innovation Groups at Yahoo, in SF, has produced. Her YHaus team has
created dozens of innovative, interactive feeds of real Yahoo!
data. These show flight segments from Y!Trip Planner turned into
dynamic parabolic trails of light, Yahoo Answers feeds showing key
words as animated word clouds, the extent of mail traffic reach and
topical search query bursts around the world. Some people have
described these applications as 'useful' art, but Joy thinks they may
be a start towards a different type of interface, an ambient one.
About the Speaker: S. Joy Mountford has been designing interfaces for
over 25 years for applications from aircraft to personal computers to
consumer devices. She has become an internationally recognized leader
in user-centered interaction design. She has led design efforts
creating interfaces to audio and visual devices, interfaces between
the electronic world and the physical world of printed materials, and
for toys, as well as for interactive music creation and generation. In
1990 she pioneered forming the Interface Design Project which sponsors
interdisciplinary design at universities around the world, and
continues to lead this effort for various sponsor companies. She
headed the Human Interface Group at Apple Computer for 8 years and
then moved to Interval Research for 6 years to lead a series of
consumer music product teams. She then led her own interaction design
company, idbias, consulting on creating interaction designs for a
range of client companies.
In 2005, Joy joined Yahoo Inc. Her initial focus was on leading a team
for re-designing for the most trafficked Web page, Yahoo's Front Page.
Currently she is leading the user experience and design efforts for
Yahoo's Communications and Community products. In addition she leads a
Design Innovation Group in San Francisco focusing on data
visualizations that are useful, as well as beautiful. Her project
interests center around building more extensive creative spaces to
position technology in an
appealing way.
____________
CS545: STANFORD-HP INFOSEMINAR
on Friday, 15 February 2008, 4:15pm - 5:30pm
Gates B03
http://infolab.stanford.edu/infoseminar/
"Towards a Distributed Web Search Engine"
Ricardo Baeza-Yates
Yahoo! Spain and Chile
http://www.dcc.uchile.cl/~rbaeza/
In the ocean of Web data, Web search engines are the primary way to
access content. As the data is on the order of petabytes, current
search engines are very large centralized systems based on replicated
clusters. Web data, however, is always evolving. The number of Web
sites continues to grow rapidly (150 millions in November 2007) and
there are currently more than 20 billion indexed pages. On the other
hand, Internet users are above one billion and hundreds of million of
queries are issued each day. In the near future, centralized systems
are likely to become less effective against such a data-query load,
thus suggesting the need of fully distributed search engines. Such
engines need to maintain high quality answers, fast response time,
high query throughput, high availability and scalability; in spite of
network latency and scattered data. In this talk we present the main
challenges behind the design of a distributed Web retrieval system and
our past and on-going work in solving some of them, including
crawling, indexing and query processing.
____________
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
sympa@lists-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). You will be asked to confirm
the subscription in either case. To unsubscribe use the word
unsubscribe instead of subscribe. Problems with subscribing or
unsubscribing should be sent to incalendar@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