[Date Prev][Date Next][Date Index]
CSLI Calendar, Wednesday, 14 November 2007, vol. 23:11
CSLI CALENDAR OF PUBLIC EVENTS
______________________________________________________________________
14 NOVEMBER 2007 Stanford Vol. 23, No. 11
______________________________________________________________________
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 11 NOVEMBER 2007 TO 23 NOVEMBER 2007
WEDNESDAY, 14 NOVEMBER 2007
12:15pm Psychology Developmental Brownbags [14-Nov-07]
Jordan Hall 420:102
"Statistical Inference and inductive learning in infants and children"
Fei Xu
University of British Columbia
http://www-psych.stanford.edu/events_developmental.html
2:00pm UC Berkeley Seminar [14-Nov-07]
207 South Hall (UC Berkeley)
"Revisiting Query Clarity:
a Distinctiveness Measure for Information Retrieval"
Daniel Tunkelang
Chief Scientist Endeca
Abstract below
3:30pm Stanford Semantics and Pragmatics Workshop [14-Nov-07]
Margaret Jacks Hall 460:126
"Force dynamics and the semantics of negative causation"
Phillip Wolff
Emory University
(cosponsored with SPLaT!; refreshments at 5:15pm)
http://www-linguistics.stanford.edu/semgroup/
Abstract below
3:45pm Psychology Department Colloquium [14-Nov-07]
Jordan Hall 420:041
"The Development of Social Categories"
Kristin Shutts
Harvard University
http://www-psych.stanford.edu/events_colloquium.html
4:00pm Berkeley School of Information Distinguished Lecture [14-Nov-07]
202 South Hall (Berkeley)
"Disruptive Innovations I Have Known and Loved -
Part 2: The Internet and the World Wide Web"
Mitch Kapor
Open Source Applications Foundation
http://www.ischool.berkeley.edu/about/events
4:10pm UC Berkeley Howison Lecture [14-Nov-07]
Toll Room, Alumni House (Berkeley)
"What We See"
Fred Dretske
Professor Emeritus, Philosophy, Stanford University
http://www.grad.berkeley.edu/lectures/howison/dretske.shtml
Abstract below
4:15pm EE380: Computer Systems Lab Colloquium [14-Nov-07]
Gates B03 (NEC Auditorium)
"On the Road to Computer Literacy"
Robert M. Lefkowitz
Asurion
http://ee380.stanford.edu/contents.html
Abstract below
6:30pm SF Bay ACM Data Mining SIG [14-Nov-07]
SAP LABS, Building D, 3410 Hillview Avenue, Palo Alto, CA
"Mapping and aligning sequences to genomes across different species"
Thomas Wu
http://www.gene.com/gene/research/sci-profiles/bioinformatics/wu/
Bioinformatics, Genentech
http://sfbayacm.org/dmsig.php
Abstract below
7:00pm SF Bay ACM TechMaster Talk [14-Nov-07]
Hewlett Packard, Pruneridge and Wolfe, Cupertino, Bldg. 48, Oak Room
"Crucial Marketing Concepts for Technology Introduction"
Sean Murphy
http://www.skmurphy.com/
http://sfbayacm.org/
Abstract below
THURSDAY, 15 NOVEMBER 2007
11:00am CCRMA Hearing Seminar [15-Nov-07]
CCRMA Seminar Room, The Knoll
"Acoustic Chord Transcription and Key Extraction from Audio
Using Key-Dependent HMMs Trained on Synthesized Audio"
Kyogu Lee
CCRMA and Gracenote
http://ccrma-mail.stanford.edu/mailman/listinfo/hearing-seminar
Abstract below
11:00am Berkeley International Computer Science Institute [15-Nov-07]
ICSI, Conference room 5A (UC Berkeley)
"Who Is Dominant?:
A Neat Solution Using A Single Distant Microphone"
Hayley Hung
IDIAP
http://www.icsi.berkeley.edu/
Abstract below
12 noon CSLI CogLunch [15-Nov-07]
Nora Suppes Hall 103
"Apprenticeship Learning with Applications in Autonomous
Helicopter Flight and Robot Dogs"
Pieter Abbeel
Stanford University
http://www-csli.stanford.edu/events/Coglunch/
Abstract below
12 noon CTL Talk [15-Nov-07]
Hartley Conference room, Mitchell Earth Sciences
"Blending Face-to-Face and Online Teaching: Successes and Challenges"
Kim McShane
The Institute for Teaching & Learning (ITL), The University of Sydney
http://events.stanford.edu/events/121/12147/
Abstract below
3:00pm SRI CCB Seminar Series [15-Nov-07]
AE201, SRI International
"Bioinformatics - an Australian perspective: through the eyes
of the Queensland Facility for Advanced Bioinformatics"
Jeremy Baker
CEO, Queensland Facility for Advanced Bioinformatics, Australia
Abstract below
4:00pm PARC Forum [15-Nov-07]
George Pake Auditorium at PARC
"Made of People"
Ross Mayfield
SocialText
http://ross.typepad.com/about.html
http://www.parc.com/forum/
Abstract below
4:00pm UC Berkeley CIS Seminar [15-Nov-07]
Soda Hall 320 (UC Berkeley)
"Learning Bounded Treewidth Bayesian Networks"
Gal Elidan
Stanford University
http://ai.stanford.edu/~galel/
http://www.cs.berkeley.edu/~rakhlin/cis-seminar
Abstract below
4:15pm SSP10: Symbolic Systems Forum [15-Nov-07]
Bldg. 380:380C (Math Corner)
"Expected Value and the Neural Prediction of Decisions"
Brian Knutson
Psychology Department, Stanford
http://symsys.stanford.edu/ssp_events
Abstract below
4:15pm Fundamental Themes in Neuroscience Seminar [15-Nov-07]
Munzer Auditorium, Beckman
"Memory and how it fails: a molecular and cellular perspective"
Alcino Silva
Neurobiology, Psychiatry and Psychology, Stanford
http://www.silvalab.org/
http://nis-seminars.stanford.edu/
Abstract below
4:15pm US-ATMC [15-Nov-07]
Skilling Auditorium
"Life Sciences and Biotech Innovation in Asia Panel"
Deepak Bangalore, President and CEO, Samixa
Benjamin Chen, Managing Director, Burrill & Company
Graeme Martin, President and CEO, Takeda Research Investment, Inc.
Mun Yew Wong Director, EDB San Francisco Center, Bio*One Singapore
http://asia.stanford.edu/
FRIDAY, 16 NOVEMBER 2007
12:30pm CS547: Human-Computer Interaction Seminar [16-Nov-07]
Gates B01
"ChucK: A Computer Music Programming Language, Designing
Instruments for Laptop Orchestras"
Ge Wang
Stanford University CCRMA
http://hci.stanford.edu/cs547/
Abstract below
12:30pm UC Berkeley HWNI Student Seminar [16-Nov-07]
101 LSA (Berkeley)
"Olfactory processing in the Drosophila brain"
Rachel Wilson
Harvard University
http://neuroscience.berkeley.edu/events/
2:15pm Center for the Explanation of Consciousness [16-Nov-07]
Cordura 100
"What We See"
Fred Dretske
Professor Emeritus, Philosophy, Stanford University
http://www.grad.berkeley.edu/lectures/howison/dretske.shtml
See Howison Lecture abstract below
3:00pm Berkeley Information Access Seminar [16-Nov-07]
107 South Hall (Berkeley)
"How Does an Online Art Community Take Shape?"
Dan Perkel
http://courses.ischool.berkeley.edu/i296a-1/f07/schedule.html
Abstract below
3:30pm Stanford Semantics and Pragmatics Workshop [16-Nov-07]
Margaret Jacks Hall 460:126
"Inversion and Paradigm Collapse in the Lower Sepik Languages,
Papua New Guinea"
William Foley
University of Sydney/Stanford
(cosponsored by the Linguistics Colloquium Committee)
http://www-linguistics.stanford.edu/semgroup/
Abstract below
4:10pm UC Berkeley Logic and the Methodology of Science [16-Nov-07]
60 Evans Hall (Berkeley)
"An Overview of Logic in Game Theory"
Eric Capuit
Computer Science, Stanford University
http://logic.berkeley.edu/colloquium.html
Abstract below
MONDAY, 19 NOVEMBER 2007
4:00pm UC Berkeley Linguistics Colloquium [19-Nov-07]
182 Dwinelle Hall (Berkeley)
Title to be announced
Mary Bucholtz
UC Santa Barbara
http://www.linguistics.ucsb.edu/faculty/bucholtz/
http://linguistics.berkeley.edu/events/
6:30pm IEEE SCV Robotics and Automation [19-Nov-07]
CMU West, Moffett Field, Mountain View
"First Steps Towards Community Robotics"
Illah R. Nourbakhsh
The Robotics Institute, Carnegie Mellon University
http://ewh.ieee.org/r6/scv/ras
Abstract below
TUESDAY, 20 NOVEMBER 2007
6:00pm MIT/Stanford Venture Lab [20-Nov-07]
Bishop Auditorium, Graduate School of Business
"Web 3.0: New Opportunities on the Semantic Web"
panel discussion
http://www.vlab.org
Information below
6:30pm IEEE SCV Communications [20-Nov-07]
505 Van Ness Ave (San Francisco)
"Universal Access to Human Knowledge:
Public Access to Digital Materials"
Brewster Kahle
Director and Founder, Internet Archive
http://www.e-grid.net/calendar.html
(rsvp required)
WEDNESDAY, 21 NOVEMBER 2007
7:30pm IEEE SCV Engineering in Medicine and Biology [21-Nov-07]
Clark Center Auditorium
"Technologies and Applications of Simulation in Healthcare &
Tour of Simulation Center at Stanford"
David M. Gaba
Stanford
http://www.ewh.ieee.org/r6/scv/embs/pages/upcoming.html
Information below
THURSDAY, 22 NOVEMBER 2007 - Thanksgiving Holiday
FRIDAY, 23 NOVEMBER 2007 - Thanksgiving Holiday
____________
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.
____________
NOTE
The Center for the Explanation of Consciousness (tentative name) is a
new group within CSLI. It is presenting Fred Dretske, Professor
Emeritus of Stanford, on Friday at 2:15pm for a talk entitled, "What
We See: The Texture of Conscious Experience". It hopes this will be
the first of a series of talks. For more information on the new
group, contact Paul Skokowski, paulsko ... csli.stanford.edu.
Stanford has no official classes next week (November 19-30) though the
university is open Monday through Wednesday.
PARC is starting a series of special forums on Going Beyond Web 2.0.
____________
UC BERKELEY SEMINAR
on Wednesday, 14 November 2007, 2:00pm
South Hall 207 (UC Berkeley)
"Revisiting Query Clarity:
a Distinctiveness Measure for Information Retrieval"
Daniel Tunkelang
Chief Scientist Endeca
For the past three decades, most research in information retrieval has
assumed a ranked retrieval model, in which a query returns a ranking
of corpus documents by their estimated relevance to this query. This
model maps to the familiar user interface of most commercial and
academic search engines.
Despite its popularity, the ranked retrieval model suffers because it
does not provide a clear split between relevant and irrelevant
documents. This weakness makes it impossible to obtain even basic
analysis of the query results, such as the number of relevant
documents, let alone a more complicated one, such as the result
quality.
In contrast, a set retrieval model partitions the corpus into two
subsets of documents: those that are considered relevant, and those
that are not. A set retrieval model does not rank the retrieved
documents; instead, it establishes a clear split between documents
that are in and out of the retrieved set. As a result, set retrieval
models enable rich analysis of query results, which can then be
applied to improve user experience.
Armed with a set retrieval framework, we revisit query clarity, an
information gain measure introduced by Cronen-Townsend and Croft in
2002 to predict the ambiguity of a query against an information
retrieval system. While Cronen-Townsend and Croft offered evidence in
support of query clarity as a measure, subsequent research by Turpin
and Hersh in 2004 showed a lack of correlation between clarity scores
and user performance.
We claim that query clarity is an effective measure, but that it needs
to be revised to leverage a set retrieval model. We present a
normalized clarity score that measures the clarity of a query result
set relative to other document sets of its size. We discuss
theoretical results about the distribution of normalized clarity,
preliminary evidence in favor of normalized clarity as a tool to
measure query quality, and applications of normalized clarity to
interactive information retrieval.
____________
STANFORD SEMANTICS AND PRAGMATICS WORKSHOP
on Wednesday, 14 November 2007, 3:30pm
Margaret Jacks Hall 460:126
http://www-linguistics.stanford.edu/semgroup/
"Force dynamics and the semantics of negative causation"
Phillip Wolff
Emory University
(cosponsored with SPLaT!; refreshments at 5:15pm)
According to process theories of causation, people represent causation
by modeling the physical and social processes that bring about
causation in the world. These theories usually require that causal
relations involve an uninterrupted chain of influences from the cause
to the effect. A key problem for this view is the phenomenon of
"negative causation." Negative causation is present when causation
occurs in the absence of a cause. We say, for example, "The absence of
nicotine causes withdrawal" or "Lack of water causes thirst." It is
also present in cases of so-called "double prevention," or, situations
where preventions are prevented, as when, for example, rescuers
prevent guards from preventing an escape and thereby cause or allow
the escape. In all cases of negative causation, there is a gap in the
chain of influences from the cause to the effect. In my talk I show
that negative causation is not, in fact, a problem for process
theories based on force dynamics. Indeed, several patterns in the
meaning of causal expressions encoding negative causation may provide
support for process approaches over competing approaches. According to
statistical, counterfactual, and logical approaches to causation,
expressions of causation involving negation and positive causation are
symmetric: for example, NOT-CAUSE --> PREVENT and PREVENT -->
NOT-CAUSE. In contrast, from a force dynamic perspective these
different expressions are often related to each other asymmetrically:
for example, NOT-CAUSE --> PREVENT, but not PREVENT --> NOT-CAUSE.
The predictions of the force dynamic approach were supported in
several experiments in which people re-expressed causal expressions
taken from the internet and described animations depicting complex
causal interactions. Because these asymmetries cannot be explained by
statistical or logical approaches, the results support the view that
causal reasoning involves simulating the actual processes that bring
about causation in the world.
____________
UC BERKELEY HOWISON LECTURE
on Wednesday, 14 November 2007, 4:10pm
Toll Room, Alumni House (Berkeley)
http://www.grad.berkeley.edu/lectures/howison/dretske.shtml
"What We See"
Fred Dretske
Professor Emeritus, Philosophy, Stanford University
We see (at least) three fundamentally different sorts of things:
objects (a tomato), properties of these objects (the tomato's size,
shape, color, orientation), and facts about them (that it is a tomato,
that it is red). I shall be concerned with only the first: our
perception of objects. I will furthermore restrict my topic by
assuming, without argument, that the objects we see, in normal
circumstances, are ordinary dry goods--tomatoes, pencils, people,
trees and houses. I am interested in how many of these objects we see
in brief, but attentive, observation. The answer to this question
tells us something important about the nature of conscious perceptual
experience.
About the Speaker: Fred Dretske specializes in epistemology and the
philosophy of mind, with an emphasis upon self-knowledge and conscious
experience. In 1994, he was awarded the Jean Nicod Prize in Paris,
which annually recognizes the contributions of a leading philosopher
of mind. Dretske is emeritus professor of philosophy at both Stanford
University and the University of Wisconsin, and has served as senior
research scholar in the philosophy department at Duke University since
1999.
____________
EE380: COMPUTER SYSTEMS LAB COLLOQUIUM
on Wednesday, 14 November 2007, 4:15pm to 5:30pm
Gates B03 (NEC Auditorium)
http://ee380.stanford.edu/contents.html
"On the Road to Computer Literacy"
Robert M. Lefkowitz
Asurion
A recurring theme in computer science is the role of computer programs
as a means of communicating algorithms between people. Computer
Literacy, then, would be defined as the ability to read and write
computer programs. We have a long way to go before we live in a
society that achieves universal computer literacy. The consideration
of this ideal is born of the realization that the Free Software
movement can only be significant in a society where the majority of
citizens are "source code" literate.
In order to identify some of the technological gaps that hinder the
implementation of universal computer literacy, this talk looks back at
the development of literacy and the associated technologies from
before Plato through Aldus Manutius and beyond.
About the speaker: Robert Lefkowitz is Vice President for Information
Architecture at Asurion. He has spent over 30 years in corporate
information technology departments working on making computer
technology more accessible to non-IT professionals. Over the years,
the names of these areas have changed -- Timesharing, Decision
Support, Distributed Computing, End-User Computing, Data Warehousing,
Business Intelligence, Content Management -- but the aspiration
remains the same. In pursuit of this goal of spreading computer
literacy, he has worked in the nuclear power, airline, investment
banking, telecommunications, and insurance industries. He holds a
degree in Computer Science from MIT.
____________
SF BAY ACM DATA MINING SIG
on Wednesday, 14 November 2007, 6:30pm - 9:00pm
SAP LABS, Building D, 3410 Hillview Avenue, Palo Alto, CA
http://sfbayacm.org/dmsig.php
"Mapping and aligning sequences to genomes across different species"
Thomas Wu
http://www.gene.com/gene/research/sci-profiles/bioinformatics/wu/
Bioinformatics, Genentech
The availability of entire genome sequences has prompted researchers
in bioinformatics to use genomes as universal coordinate systems for
mapping various types of sequence information. In turn, bioinformatics
researchers need computer programs that can efficiently and accurately
map and align nucleotide and protein sequences to entire genomes. The
problem of mapping and aligning cDNA sequences to genomes for the same
species has essentially been solved, with programs such as GMAP. The
remaining issues in the field are the ability to handle cross-species
alignments and protein sequences. Cross-species alignments enable us
to leverage our knowledge about one species onto another. Mapping and
alignment of protein sequences has certain advantages over nucleotide
sequences, especially in cross-species comparisons, because protein
sequences are evolutionarily more conserved than nucleotide
sequences. In this talk, I will discuss our recent work in extending
the algorithmic ideas of GMAP to handle cross-species and protein
sequences.
About the Speaker: Thomas Wu received his B.S. and M.S. from Stanford
in 1984, his M.D. from Harvard Medical School in 1992, and his
Ph.D. in Computer Science from MIT in 1992. He completed his
internship and residency in Internal Medicine at Stanford in 1994. He
was a Postdoctoral Fellow in the Department of Biochemistry at
Stanford from 1994 to 1999. Since 1999, Dr. Wu has been a Senior
Scientist in the Department of Bioinformatics at Genentech.He is a
independent consultant and currently the Analytics Evangelist for
Google. Prior to that he was the Director of Web Research & Analytics
for Intuit where he was responsible for the business, technical and
strategic elements of the analytics platform that supported more than
70 Intuit websites.
____________
SF BAY ACM TECHMASTER TALK
on Wednesday, 14 November 2007, 6:30pm - 9:00pm
Hewlett Packard, Pruneridge and Wolfe, Cupertino, Bldg. 48, Oak Room
http://sfbayacm.org/
"Crucial Marketing Concepts for Technology Introduction"
Sean Murphy
http://www.skmurphy.com/
Spend an hour and leave with a summary of key marketing insights, some
rules of thumb for successful innovation, and an explanation of the
evolution of technology markets. You will gain a fundamental
understanding of the difference between selling an invention and
marketing an innovation. Sean will cover strategies that form the
basis for conventional wisdom in Silicon Valley for marketing
discontinuous or disruptive products.
About the Speaker: Sean Murphy has taken an entrepreneurial approach
to life since he could drive. He has served as an adviser to dozens of
startups, helping them explore new options and bring their businesses
to new levels. His firm, SKMurphy, Inc., focuses on early customers
and early revenue for software startups, helping engineers to
understand business development.
Prior to SKMurphy, Sean worked in a variety of areas including
software engineering, engineering management, application engineering,
business development, product marketing and customer support. His
clients include Cisco Systems, 3Com, AMD, MMC Networks, Escalade and
VLSI Technology. Sean holds a BS in Mathematical Sciences and an MS in
Engineering-Economic Systems from Stanford University.
____________
CCRMA HEARING SEMINAR
on Thursday, 15 November 2007, 11:00am
CCRMA Seminar Room, The Knoll
http://ccrma-mail.stanford.edu/mailman/listinfo/hearing-seminar
Last week Elaine Chew gave a really wonderful description of her
approach for chord recognition and visualization. It was based on
transforming the notes into a special musical representation so that
triangular connections can be used with simple nearest-neighbor
decisions to recognize the current chord. It was very beautiful.
Kyogu Lee has been working on a similar problem, but taking an
entirely different approach. While Elaine's approach is driven by
music knowledge, Kyogu's approach is entirely driven by machine
learning. Give him lots of data and his system will learn the musical
knowledge to decide the chord. These data-driven algorithms are the
reason that speech recognition works. As they say in the
speech-recognition world: there is no data like more data.
In addition, Kyogu has taken the unique approach by starting with
symbolic audio. He can use symbolic means to recognize the chords and
the keys from the notes. And then use the same symbolic music to
synthesize audio. The combination allows him to learn a rich model
for each chord in each key.
"Acoustic Chord Transcription and Key Extraction from Audio
Using Key-Dependent HMMs Trained on Synthesized Audio"
Kyogu Lee
CCRMA and Gracenote
Automatically transcribing chords and extracting keys from musical
audio is a challenging task. In this talk, I will describe an acoustic
chord transcription system that uses symbolic data to train hidden
Markov models and gives best-of-class frame-level recognition
results. We avoid the extremely laborious task of human annotation of
chord names and boundaries - which must be done to provide machine
learning models with ground truth - by performing automatic harmony
analysis on symbolic music files. In parallel, we synthesize audio
from the same symbolic files and extract acoustic feature vectors
which are in perfect alignment with the labels. We, therefore,
generate a large set of labeled training data with a minimal amount of
human labor. This allows for richer models. Thus, we build 24
key-dependent HMMs, one for each key, using the key information
derived from symbolic data. Each key model defines a unique
state-transition characteristic and helps avoid confusions seen in the
observation vector.
Given acoustic input, we identify a musical key by choosing a key
model with the maximum likelihood, and we obtain the chord sequence
from the optimal state path of the corresponding key model; both of
which are returned by a Viterbi decoder. This not only increases the
chord recognition accuracy, but also gives key information.
Experimental results show the models trained on synthesized data
perform very well on real recordings, even though the labels
automatically generated from symbolic data are not 100% accurate. We
also demonstrate the robustness of the tonal centroid feature, which
outperforms the conventional chroma feature.
About the Speaker: Kyogu Lee received the B.S. degree in Electrical
Engineering from Seoul National University, Seoul, Korea, in 1996, the
M.M. degree in Music Technology from New York University, New York, in
2002, and the M.S. degree in Electrical Engineering from Stanford
University, Stanford, in 2007. He is expecting the Ph.D. degree in
Computer-Based Music Theory and Acoustics at the Center for Computer
Research in Music and Acoustics (CCRMA) at Stanford University in
January 2008. He is currently a Research Engineer at Gracenote. His
research interests include the application of signal processing and
machine learning techniques towards music and multimedia information
retrieval.
____________
BERKELEY INTERNATIONAL COMPUTER SCIENCE INSTITUTE
on Thursday, 15 November 2007, 11:00am
Conference room 5A, ICSI, 1947 Center Street, Berkeley
http://www.icsi.berkeley.edu/
"Who Is Dominant?:
A Neat Solution Using A Single Distant Microphone"
Hayley Hung
IDIAP
In this talk, I will describe some joint work between ICSI and IDIAP
on applying different speaker diarization strategies from a single
source for the task of estimating the dominant person in a group
meeting. Previous work has shown that speaking length is strongly
correlated with perceived dominance. Here we investigate this in more
depth by considering other dominance tasks where there is full and
majority agreement amongst ground-truth annotators. In addition, we
investigate how 24 different speed-up and algorithmic strategies, and
source types lead to interesting outcomes when applied to dominance
estimation. We obtained the best performance of 77% using our slowest
scheme and a single distant microphone (SDM). Within the top 3 out of
24 performing experiments in both dominance tasks, we show that we can
use the furthest SDM, with no prior knowledge of the number of
speakers and the fastest diarization scheme, which performs 1.3 times
faster than real-time.
About the Speaker: Hayley Hung is a Post-doctoral researcher at the
IDIAP research institute in Switzerland, working on the ROADMAP
project which is a joint venture between ICSI and IDIAP. She obtained
her PhD degree in computer vision from Queen Mary University of
London, supervised by Prof. Shaogang Gong, which was part sponsored by
QinetiQ Ltd. She also has an MEng degree in Electronic and Electrical
Engineering from Imperial College, London. Her research interests
covers areas of multi-modal research related to human behaviour
understanding or inference.
____________
CSLI COGLUNCH
on Thursday, 15 November 2007, 12 noon - 1:00pm
Nora Suppes Hall 103
http://www-csli.stanford.edu/events/Coglunch/
"Apprenticeship Learning with Applications in Autonomous
Helicopter Flight and Robot Dogs"
Pieter Abbeel
Stanford University
Many problems in control have unknown, stochastic, and highly
non-linear dynamics, and offer significant challenges to classical
control methods. Some of the key difficulties in these problems are
that (i) It is often hard to write down, in closed form, a formal
specification of the control task (for example, what is the objective
function for "driving well"?), (ii) It is difficult to learn good
control---as opposed to merely descriptive---models of the dynamics
(cf. the "exploration problem" in reinforcement learning), and (iii)
It is expensive to find closed-loop controllers for high dimensional,
highly stochastic domains. In this talk, I will present formal results
showing how these problems can be efficiently addressed in the
apprenticeship learning setting, in which expert demonstrations of the
task are available. I will also present an application of our ideas to
autonomous helicopter flight and four-legged locomotion. Our
helicopter results significantly extend the state of the art in
helicopter control, and include the first successful completion of the
following five aerobatic flight maneuvers: tic-toc, in-place forward
flip and sideways roll, nose-in funnel, and tail-in funnel. Our
resulting robot dog controller constitutes the state of the art in
quadruped locomotion across rugged, previously unseen terrains.
About the Speaker: Pieter Abbeel is a PhD student in Prof. Andrew Ng's
group at Stanford University. His research interests include machine
learning, robotics, and control.
____________
CTL TALK
on Thursday, 15 November 2007, 12 noon
Hartley Conference room, Mitchell Earth Sciences
http://events.stanford.edu/events/121/12147/
"Blending Face-to-Face and Online Teaching: Successes and Challenges"
Kim McShane
The Institute for Teaching & Learning (ITL), The University of Sydney
Drawing on examples from her recent research that investigated faculty
experiences of making the move online, Dr. McShane will outline some
of the successes and challenges associated with online and blended
teaching. There will be an opportunity for all to share examples of
what they know does work well online and face-to-face. In these times
of student-centred learning is it useful to reflect on what is
learning and what it is becoming in our universities and colleges? In
this presentation, Dr. McShane will review how online learning might
be changing university teaching and, indeed, changing student
learning, in significant ways. When distance learning is not the only
option, what case can we make for taking up online teaching, and what
case do we make for integrating face-to-face and online teaching?
Cosponsored by the Center for Teaching and Learning and the Stanford
Center for Professional Development
____________
SRI CCB SEMINAR SERIES
on Thursday, 15 November 2007, 3:00pm
AE201, SRI International
"Bioinformatics - an Australian perspective: through the eyes
of the Queensland Facility for Advanced Bioinformatics"
Jeremy Baker
CEO, Queensland Facility for Advanced Bioinformatics, Australia
Australia covers a large geographic area with large distances between
the main research centres. Because of this, bioinformatics capability
has grown from those groups that started out in genomics research
largely independently. As often happens in the domain, there is a lot
of replication of capability and skills with little cohesion of
effort. Bioinformatics groups in Brisbane, Queensland recognised that
there was no group providing local up-to-date data which they needed
for their projects. Hence the genesis of the Queensland Facility for
Advanced Bioinformatics, (QFAB). Together, the QFAB collaborators and
the Queensland State government have funded the initiative for three
years. QFAB aims to deliver access to data as well as bioinformatics
capability to researchers throughout Australia. Initially it is
focussed on delivering against its partner's requirements and will
soon begin some challenging national projects. This talk will outline
some of the QFAB and Australian bioinformatics related infrastructure
and projects that rely on it.
About the Speaker: Mr Barker joined the Queensland Facility for
Advanced Bioinformatics as the full time CEO in June 2007 having
previously been contracted to arrange the planning, personnel, finance
and contractual arrangements for the Facility.
Mr Barker has a track record of establishing or growing a number of
organisations in the life sciences sector. These have included high
technology marine production facilities, a biotechnology advisory
business, seven years developing the Australian Genome Research
Facility Ltd which includes a specialist bioinformatics service, and
is a director and shareholder in ESPData Pty Ltd which provides
pre-clinical and clinical trials capability.
Mr Barker originally graduated with a Bachelor of Science and Masters
of Science from Auckland University, New Zealand. His thesis covered
the population genetics of a freshwater fish complex using allozyme
electrophoresis. He also holds an MBA from Deakin University,
Australia and is a Member of the Australian Institute of Company
Directors and a Fellow of the Australian Institute of Management. He
has been actively involved as member and Chair of the Queensland
Branch of Ausbiotech Ltd, the peak body for Australian biotechnology.
____________
PARC FORUM
on Thursday, 15 November 2007, 4:00pm - 5:00pm
George Pake Auditorium, PARC
http://www.parc.com/forum/
"Made of People"
Ross Mayfield
SocialText
http://ross.typepad.com/about.html
All things 2.0 are made of people. The social software that powers the
current wave of innovation takes a different approach of getting out
of the way of people to unleash their abundant desire to share and
collaborate. While these tools exhibit fantastic social dynamics on
the public web, adapting them for the context of an organization is a
challenge not only for tools, but practices. Sharing control to create
value isn't exactly the instinct of the enterprise. This talk will
explore the social software design and business patterns that might
make us more human.
About the Speaker: Ross Mayfield is the Chairman, President and
Co-founder of Socialtext, the first wiki company and leading provider
of Enterprise 2.0 solutions. A well known blogger and speaker on web
trends, you can find him at http://ross.typepad.com
____________
UC BERKELEY CIS SEMINAR
on Thursday, 15 November 2007, 4:00pm-5:00pm
Soda Hall 320 (UC Berkeley)
http://www.cs.berkeley.edu/~rakhlin/cis-seminar
"Learning Bounded Treewidth Bayesian Networks"
Gal Elidan
Stanford University
http://ai.stanford.edu/~galel/
With the increased availability of real data for complex domains, it
is particularly appealing to automatically learn Bayesian network
structures that are sufficiently expressive but where inference is
still tractable. While the method of thin junction trees can, in
principle, be used for this purpose, its fully greedy nature makes it
somewhat prone to over-fitting, particularly when the data is sparse
and when the desired treewidth is not trivial.
In this talk I will present a novel method for efficiently learning
Bayesian networks of bounded treewidth that employs global structure
modifications. At the heart of our method is the idea of dynamically
updating a triangulation in a way that facilitates the addition of
optimal chain structures (with respect to some ordering) that are
guaranteed to increase the treewidth bound of the model by at most
one. We demonstrate the effectiveness of our "treewidth friendly"
method on real-life datasets and show that it is superior to the
greedy approach as soon as the bound on the treewidth is nontrivial.
Importantly, we show that by employing global operators that are
closer in spirit to those used by Chow and Liu for learning optimal
trees, we are able to learn better networks even when the treewidth of
the model is not bounded.
____________
SSP10: SYMBOLIC SYSTEMS FORUM
on Thursday, 15 November 2007, 4:15pm
Bldg. 380:380C (Math Corner)
http://symsys.stanford.edu/ssp_events
"Expected Value and the Neural Prediction of Decisions"
Brian Knutson
Psychology Department, Stanford
The past decade has introduced revolutionary advances in scientists'
understanding of the neural mechanisms that support human decision
making, in part due to advances in the spatiotemporal resolution of
brain imaging. I will describe brain imaging research from our
laboratory designed to: (1) localize brain regions whose activation
correlates with expected value in the absence of choice, and (2) use
activation from those regions to predict choice. Such research may
have practical applications to the study of financial risk taking and
purchasing. The findings also theoretically imply that a revealed
preference account of human decision making, while useful, is
incomplete.
____________
FUNDAMENTAL THEMES IN NEUROSCIENCE SEMINAR
on Thursday, 15 November 2007, 4:15pm
Munzer Auditorium, Beckman
http://nis-seminars.stanford.edu/
"Memory and how it fails: a molecular and cellular perspective"
Alcino Silva
Neurobiology, Psychiatry and Psychology, Stanford
http://www.silvalab.org/
Our laboratory is studying the role of hippocampal and prefrontal
function in recent and remote memory. We are interested in uncovering
general rules for how these two brain regions account for their unique
roles in recent and remote memory. We are also studying how molecular
mechanisms modulate cellular responses that underlie key microcircuit
properties implicated in learning and memory. To accomplish this we
are using a myriad of techniques including region and temporal
specific transgenic manipulations, viral vectors, pharmacology, in
vivo and in vitro electrophysiology, in vivo two-photon scanning
confocal microscopy, as well as a battery of behavioral tasks designed
to probe hippocampal and prefrontal cortical function. Our laboratory
is also interested in understanding the mechanisms underlying
cognitive disorders such as those underlying learning disabilities
associated with Neurofibromatosis Type 1 (NF1), Tuberous Sclerosis and
schizophrenia.
____________
CS547: HUMAN-COMPUTER INTERACTION SEMINAR
on Friday, 16 November 2007, 12:30-2:00pm
Gates B01
http://hci.stanford.edu/cs547/
"ChucK: A Computer Music Programming Language, Designing
Instruments for Laptop Orchestras"
Ge Wang
Stanford University CCRMA
http://chuck.cs.princeton.edu/
http://plork.cs.princeton.edu/
In the first part of this talk, we present the design, philosophy, and
development of ChucK, a computer music programming language intending
to provide a different approach, expressiveness, and thinking with
respect to time and parallelism in audio programming - as well as a
platform for precise and rapid experimentation. The basic tenets of
ChucK include a syntax for representing audio flow, a new time-based
programming model that allows programmers to precisely control time
across concurrent program components (we call this "strongly-timed"),
and facilities to rapidly experiment with programs "on-the-fly" (i.e.,
as they run). A ChucKian approach to "live coding" as a new musical
performance paradigm is also discussed. This in turn motivates the
Audicle: a graphical environment to visualize audio programming in
real-time. We also present the applications of ChucK in audio
research, composition/performance, and education.
In the second part of this presentation, we describe our adventures
with the "laptop orchestra": a new type of large-scale,
computer-mediated music ensemble. The laptop orchestra consists of 12
or more sets of laptops, humans, special hemispherical speakers,
sensors, and software, and presents new challenges in music
technology, instrument design, composition, performance, and
pedagogy. Since its instantiation at Princeton University in 2005 (as
the Princeton Laptop Orchestra, or PLOrk), the orchestra has premiered
more than 50 new compositions and offered four new courses (in
Computer Science and Music), using ChucK as a primary tool for
teaching, composition, and instrument design in the ensemble and
classroom. In these contexts, we present our ongoing experiences in
building new human-computer musical instruments and performances, and
discuss the potential of the laptop orchestra as a unique platform for
teaching and experimentation with music and technology. Lastly, we
present plans for the upcoming "Stanford Laptop Orchestra".
About the Speaker: Ge Wang received his B.S. in Computer Science in
2000 from Duke University, PhD (soon) in Computer Science (advisor
Perry Cook) in 2007 from Princeton University, and is currently an
assistant professor in the Center for Computer Research in Music and
Acoustics (CCRMA) at Stanford University. His research interests
include interactive software systems (of all sizes) for computer
music, programming languages, sound synthesis and analysis, music
information retrieval, new performance ensembles (e.g., laptop
orchestra) and paradigms (e.g., live coding), visualization,
interfaces for human-computer interaction, interactive audio over
networks, and methodologies for education at the intersection of
computer science and music. Ge is the chief architect of the ChucK
audio programming language and the Audicle environment. He is a
founding developer and co-director of the Princeton Laptop Orchestra
(PLOrk), and a co-creator of the TAPESTREA sound design
environment. Ge composes and performs via various electro-acoustic and
computer-mediated means, including with PLOrk, with Perry as a live
coding duo, and with Princeton graduate student and comrade Rebecca
Fiebrink in a duo exploring new performance paradigms, cool audio
software, and great food.
Ge joins the Stanford music faculty as an assistant professor in the
Center for Computer Research in Music and Acoustics (CCRMA). He
excitedly looks forward to many things, including working with the
strong faculty, researchers, and students and forging new directions
in computer music research, software, and pedagogy, as well as writing
music and initiating and developing new performance ensembles and
paradigms, including a Stanford Laptop Orchestra.
____________
BERKELEY INFORMATION ACCESS SEMINAR
on Friday, 16 November 2007, 3:00pm - 5:00pm
107 South Hall (Berkeley)
http://courses.ischool.berkeley.edu/i296a-1/f07/schedule.html
"How Does an Online Art Community Take Shape?"
Dan Perkel
How do ongoing practices of making and sharing art shape the
development of the technological aspects of an art community website
and how does the production of the site in turn shape the social and
creative practices of the diverse groups of people involved? What are
the various ways in which the site's "users" influence the
technological production of the site? How is the creative expression
and creative production going on in the site structured by the social
organization, technical implementation, and ongoing social practices
on the site? Finally, what role do youth play in the site's
development and how do they understand their own practices on the
site? I intend an ethnographic approach to technology use and design
by participating on the site, finding ways to get into the lives and
activity of the site's members (including both "users" and
"designers," a highly problematic distinction on this particular
site), and tracing the historical and ongoing development of
particular site features and practices from multiple points of view.
____________
STANFORD SEMANTICS AND PRAGMATICS WORKSHOP
on Friday, 16 November 2007, 3:30pm
Margaret Jacks Hall 460:126
http://www-linguistics.stanford.edu/semgroup/
"Inversion and Paradigm Collapse in the Lower Sepik Languages,
Papua New Guinea"
William Foley
University of Sydney/Stanford
(cosponsored by the Linguistics Colloquium Committee)
Direct versus inverse inflectional systems are a commonplace feature
of languages which signal their grammatical relations primarily by
verbal agreement, such as many Amerindian languages or Tibeto-Burman
languages of the Himalayas. Direct versus inverse systems are
characterized by one inflectional pattern when a speech act
participant or local person (first or second person) functions as
actor to a non-speech act participant or nonlocal person (third
person) as patient and another inflectional pattern when the opposite
situation holds. Scenarios in which local participants (first and
second person) act on each other pose particular problems. The
Algonkian languages of North America are the paradigm case of this
grammatical inflectional pattern. As the Lower Sepik languages are
morphologically complex languages which express grammatical
information almost exclusively through verbal morphology, they, not
unexpectedly, exhibit direct-inverse inflectional systems, and such a
system was unquestionably a feature of Proto-Lower Sepik. However,
while all six currently extant languages have such systems, each is
different to a greater or lesser extent from the others, particularly
in dealing with the pragmatically complex scenario in which local
participants act on each other. I will employ a version of Optimality
Theory to set out an analysis of the parameters of variation across
the languages. Starting with a revision of Wunderlich's revision of my
description of Yimas, I will propose 5 constraints that determine the
patterns of verb inflection for transitive verbs in that language. I
will then look at each language in turn and demonstrate how the
variation among them is due to different ranking of these constraints,
and in some cases, due to grammatical changes elsewhere, their
apparent loss. The findings have important implications for procedures
for the reconstruction of morphological systems in comparative
linguistics, but also, due to the high degree of homophony in the
paradigms in some languages, particularly in the inverse forms, and
the disjunctive semantics of a number of the morphemes, for
morphological theory more generally.
____________
UC BERKELEY LOGIC AND THE METHODOLOGY OF SCIENCE
on Friday, 16 November 2007, 4:10pm - 5:30pm
60 Evans Hall (Berkeley)
http://logic.berkeley.edu/colloquium.html
"An Overview of Logic in Game Theory"
Eric Capuit
Computer Science, Stanford University
Game-theoretic methods have proven to be an important tool in logic.
In this talk, I will present a different perspective on the interface
between logic and game theory: how logic can be used to reason about
social interactive situations. This will be a survey talk focusing
on:
(1) how logical methods provide interesting new perspectives on
traditional game-theoretic questions, and
(2) new questions that arise of interest to both game-theorists and
logicians.
____________
IEEE SCV ROBOTICS AND AUTOMATION
on Monday, 19 November 2007, 6:30pm
CMU West, Moffett Field, Mountain View
http://ewh.ieee.org/r6/scv/ras
"First Steps Towards Community Robotics"
Illah R. Nourbakhsh
The Robotics Institute, Carnegie Mellon University
The CREATE lab and the Global Connection Project have embarked on a
series of public projects to try and understand how significant
scaling may be feasible using robotics for technology empowerment and
community-building. Our work is now hybridizing the Global Connection
efforts together with our more traditional Telepresence Robot Kit and
CMUcam educational tools, and we are carrying out experiments locally
in Pittsburgh and internationally in collaboration with UNESCO. I will
describe the current status of our community products, describing both
our target communities spanning the cognitive pipeline, and the new
technologies we are releasing this year (CMUcam3, a fully
programmable, public-domain embedded computer vision system; TeRK, a
single-box solution for complete I/O control of robots together with a
Linux OS and connectivity to the iRobot Create and other robots;
Canary, a new embedded environmental sensor and kinetic art
controller; and Gigapan, a multi-billion pixel panoramic image
capture, display and annotation system. I will end by describing the
beginning of our unified effort at large-scale public, activist
robotic art across Pittsburgh, called Robot 250, which is funded by a
group of five corporations and foundations. The Global Connection
Project is hosted by Carnegie Mellon West (Moffett Field, CA) and
Carnegie Mellon Robotics Institute (Pittsburgh, PA).
About the Speaker: Illah R. Nourbakhsh is an Associate Professor of
Robotics and head of the Robotics Masters Program in The Robotics
Institute at Carnegie Mellon University. He was on leave for the 2004
calendar year and was at NASA/Ames Research Center serving as Robotics
Group lead. He received his Ph.D. in computer science from Stanford
University in 1996. He is co-founder of the Toy Robots Initiative at
The Robotics Institute, director of the Center for Innovative Robotics
and director of the Community Robotics, Education and Technology
Empowerment (CREATE) lab. He is also co-PI of the Global Connection
Project, home of the Gigapan project. He is also co-PI of the Robot
250 city-wide art+robotics fusion program in Pittsburgh. His current
research projects include educational and social robotics and
community robotics. His past research has included protein structure
prediction under the GENOME project, software reuse, interleaving
planning and execution and planning and scheduling algorithms, as well
as mobile robot navigation. At the Jet Propulsion Laboratory he was a
member of the New Millennium Rapid Prototyping Team for the design of
autonomous spacecraft. He is a founder and chief scientist of Blue
Pumpkin Software, Inc., which was acquired by Witness Systems,
Inc. Illah recently co-authored the MIT Press textbook, Introduction
to Autonomous Mobile Robots.
____________
MIT/STANFORD VENTURE LAB
on Tuesday, 20 November 2007, 6:00pm
Bishop Auditorium, Graduate School of Business
http://www.vlab.org
"Web 3.0: New Opportunities on the Semantic Web"
panel discussion
We are well into the current era of the Web, commonly referred to as
Web 2.0.
What lies on the horizon? Will Web 3.0 usher in the long awaited
vision of the semantic web, as proposed by "Father of the Web" Tim
Berners-Lee more than ten years ago?
Join us for a lively panel session where some of the best emerging
companies in the semantic web space present their different approaches
to realizing the vision. The panel will address questions such as: How
can we best implement the vision of the semantic web? What will we do
with the web once it is structured with semantic information? What new
applications will appear? Where is the consumer value and how should
it be marketed? What new businesses can be built on top of the
semantic web that are not possible today? Will the semantic web
ultimately bring about a new intelligence that surpasses that of
humanity, sparking a new era of non-biological evolution?
Moderator:
Paul Saffo, Technology Forecaster and Consulting Professor, Stanford
Panelists:
Robert Cook, Co-founder and Executive VP of Product Development, Metaweb
Nova Spivack , CEO and Founder, Radar Networks
Alex Iskold , CEO and Founder, Adaptive Blue
Paul Kedrosky , Venture Partner, Ventures West
Registration and fee.
Regular Stanford Student (but reception costs $15)
early $35 free
after November 19 $40 free
____________
IEEE SCV ENGINEERING IN MEDICINE AND BIOLOGY
on Wednesday, 21 November 2007, 7:30pm
Clark Center Auditorium
http://www.ewh.ieee.org/r6/scv/embs/pages/upcoming.html
"Technologies and Applications of Simulation in Healthcare &
Tour of Simulation Center at Stanford"
David M. Gaba
Stanford
The last decade has seen explosive growth of technologies and
applications of simulation in healthcare. This growth was based on a
number of prior, often independent, developments of device and
curricula. There is now a wide diversity of simulation methodologies
ranging from low tech ones (verbal simulation, role-playing, using
food products -- such as beef hearts -- for practice) to higher tech
ones (part-task trainers, computerized mannequins) to truly high-tech
ones (virtual reality). These have been linked to an equally wide
diversity of applications. In fact the scope of simulation in
healthcare can be described according to (at least) 11 dimensions.
This talk will provide an overview of the modalities and applications
of simulation. There will be a brief review of some of the history,
especially that of my own laboratory's pioneering developments dating
back to 1986. Organizational aspects of the diffusion of these
technologies and applications will be discussed, including why
healthcare is different from the analogous industries that have been
used as models of simulation (aviation, nuclear power, spaceflight,
military).
There will also be a detailed tour of the new Goodman Simulation
Center (GSC) at Stanford. This center is the newest of 3 dedicated
simulation centers at Stanford and its associated hospitals. Stanford
is one of the world leaders of all varieties of simulation in
healthcare, and the GSC has many state-of-the-art (or better)
characteristics. It is also the pathfinder for a 28,000 square foot
Immersive Learning Center that will occupy the entire basement of the
new medical education building -- the Learning and Knowledge Center --
currently in the final stages of design for construction and opening
by spring 2010.
About the Speaker: David M. Gaba, M.D. is Associate Dean for Immersive
and Simulation-based Learning and Professor of Anesthesia (with
tenure) at Stanford University School of Medicine. He is also Director
of the Patient Safety Culture Institute and the Patient Simulation
Center of Innovation at Veterans Affairs Palo Alto Health Care System
where he is a Staff Anesthesiologist. Over the last 22 years
Dr. Gaba's laboratory has worked extensively on human performance and
patient safety issues. This laboratory is a pioneer in applying
organizational safety theory to health care - including both Normal
Accidents theory and High Reliability Organization theory. Dr. Gaba
and his team is the inventor of the modern full-body patient simulator
and is responsible for introducing Crew Resource Management training
from aviation to healthcare, first in anesthesia and then to many
other healthcare domains. He has been the principal investigator on
grants from a wide variety of funders, and is currently the PI on
projects concerning safety culture in hospitals and on applying
simulation to address safety culture in diverse types of hospitals
ranging from rural critical access hospitals to large urban academic
centers. Many of Dr. Gaba's fellows, faculty collaborators, and protgs
have gone on to leadership positions on human performance in
healthcare, organizational safety, and simulation in healthcare
throughout the world.
____________
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