[Date Prev][Date Next][Date Index]
CSLI Calendar, Wednesday, 4 April 2007, vol. 22:29
CSLI CALENDAR OF PUBLIC EVENTS
______________________________________________________________________
4 April 2007 Stanford Vol. 22, No. 29
______________________________________________________________________
A weekly publication of the
Center for the Study of Language and Information (CSLI)
Stanford University, Cordura Hall, Stanford, CA 94305-4101
http://www-csli.stanford.edu/
____________
ACTIVITIES FROM 4 APRIL 2007 TO 15 APRIL 2007
WEDNESDAY, 4 APRIL 2007
4:00pm UC Berkeley Psychology Talk [4-Apr-07]
3105 Tolman (Berkeley)
"The Gist of a Scene"
Aude Oliva
MIT
http://web.mit.edu/bcs/people/oliva.shtml
http://psychology.berkeley.edu/news/colloquia.html
4:15pm EE380: Computer Systems Lab Colloquium [4-Apr-07]
Gates B01 (HP Auditorium)
"Nanomanufacturing Technologies:
Extending the Silicon Roadmap and Enabling New Applications"
Mark Pinto
Applied Materials, Solar Business Group
http://ee380.stanford.edu/contents.html
5:00pm Stanford Challenge on K-12 Education Talk [4-Apr-07]
Cubberley Auditorium
"Teaching Science: How, What, and Who Decides?"
Donald Kennedy
Editor-in-Chief of Science
http://ed.stanford.edu/
Abstract below
5:30pm Tanner Lecture: Racial Stigma, Mass Incarceration and American Values
Levinthal Hall, Stanford Humanities [4-Apr-07]
Lecture 1: "Ghettos, Prisons and Racial Backlash"
Glenn Loury
Economics, Brown University
http://www.stanford.edu/dept/EIS/lectures_tanner.html
THURSDAY, 5 APRIL 2007
10:00am Tanner Lecture discussion seminar [5-Apr-07]
Landau Economics, SIEPR A
Glenn Loury
Economics, Brown University
http://www.stanford.edu/dept/EIS/lectures_tanner.html
4:00pm PARC Forum [5-Apr-07]
George Pake Auditorium at PARC
"Forensic Science and Weapons of Mass Destruction"
John G. Reynolds
Forensic Science Center, Lawrence Livermore National Laboratory
http://www.parc.com/forum/
4:15pm SSP10: Symbolic Systems Forum [5-Apr-07]
Bldg. 380:380C (Math Corner)
"Reward, Value and Choice: A Perspective on the Neurobiology
of Decision Making"
Bill Newsome
Neurobiology, Stanford
http://monkeybiz.stanford.edu/
http://symsys.stanford.edu/ssp_events
Abstract below
5:30pm Tanner Lecture: Racial Stigma, Mass Incarceration and American Values
Levinthal Hall, Stanford Humanities [5-Apr-07]
Lecture 2: "Social Identity & the Ethics of Punishment"
Glenn Loury
Economics, Brown University
http://www.stanford.edu/dept/EIS/lectures_tanner.html
FRIDAY, 6 APRIL 2007
10:00am Tanner Lecture discussion seminar [6-Apr-07]
Landau Economics, SIEPR A
Glenn Loury
Economics, Brown University
http://www.stanford.edu/dept/EIS/lectures_tanner.html
11:00am Berkeley Institute of Cognitive and Brain Seminar [6-Apr-07]
Tolman 5101 (Berkeley)
"Differentiation and Disintegration of Conceptual Knowledge:
A Parallel-Distributed Processing Approach"
Jay McClelland
Psychology, Stanford
http://icbs.berkeley.edu/
Abstract below
12:30pm CS547: Human-Computer Interaction Seminar [6-Apr-07]
Gates B03 (NEC Auditorium)
"Paying Attention to Interruption:
A Human-Centered Approach to Intelligent Interruption Management"
Brian Bailey
University of Illinois (UIUC)
http://www-faculty.cs.uiuc.edu/%7Ebpbailey/
http://hci.stanford.edu/cs547/
Abstract below
1:30pm Berkeley International Computer Science Institute [6-Apr-07]
ICSI, Rm 607 (UC Berkeley)
"Ensembles of classifiers"
Alberto Suarez
Universidad Autonoma de Madrid and ICSI
http://www.icsi.berkeley.edu/
Abstract below
3:00pm Berkeley Information Access Seminar [6-Apr-07]
107 South Hall (Berkeley)
"Objects with Narrative Memories"
Kimiko Ryokai
new joined to School of Information, Berkeley
http://web.media.mit.edu/%7Ekimiko/
http://www.ischool.berkeley.edu/about/news/topstories/ryokai10112006a
http://courses.ischool.berkeley.edu/i296a-1/s07/schedule.html
Abstract below
3:15pm Friday Cognitive Seminar [6-Apr-07]
Jordan Hall 420:050
"Neural measures of individual differences in working memory
and declarative memory: Examining the relationship between
adaptation and selectivity in object and face processing using
human and monkey fMRI"
Ben Hutchinson & Kevin Weiner
http://www-psych.stanford.edu/events_cognitive.html
3:15pm ME396: Design and Manufacturing Forum [6-Apr-07]
Thornton 102 (across pond from Terman)
"Walking and Jumping Humanoid Robot"
Trevor Blackwell
Anybots
http://anybots.com/
http://cs.stanford.edu/calendar/abstract.php?eventId=2415
(public invited to this session)
3:30pm Linguistics Department Colloquium [6-Apr-07]
Margaret Jacks Hall 460:126
"Tattered Syntaxes: Beckett's Late Style"
Ann Banfield
UC Berkeley
http://www-linguistics.stanford.edu/colloq/
Abstract below
3:30pm SRI CSL Seminar Series [6-Apr-07]
EK255, SRI International
"Overview of Real-Time Maude and its Applications"
Peter Olveczky
University of Oslo
http://www.csl.sri.com/
Abstract below
4:15pm CSLI Seminar on Computational Learning and Adaptation [6-Apr-07]
Cordura Hall 100
"Distributed Detection and Inference:
An adaptive anomaly detector for worm detection"
John Mark Agosta
Machine Learning Group, Intel Research
http://cll.stanford.edu/scla/schedule.shtml
Abstract below
MONDAY, 9 APRIL 2007
3:00pm Stanford Software Seminar [9-Apr-07]
Gates 104
"Abstract State Matching and Symbolic Execution for Software
Model Checking"
Corina Pasareanu
NASA Ames
http://theory.stanford.edu/~mhn/sss.html
4:00pm UC Berkeley Ear Club [9-Apr-07]
3105 Tolman Hall (Berkeley)
Music and mathematics"
Gareth Loy
Berkeley
http://ear.berkeley.edu/ear-club-schedule.html
4:10pm UC Berkeley 19th Annual Alfred Tarski Lecture 1 [9-Apr-07]
60 Evans Hall (Berkeley)
"Interpretations of Set Theory in Discrete Mathematics and
Informal Thinking: Interpretations, According to Tarski"
Harvey M. Friedman
Mathematics, Computer Science, and Philosophy, Ohio State University
http://logic.berkeley.edu/tarski-lectures.html
4:15pm CS528: Broad Area Colloquium [9-Apr-07]
Hewlett Teaching Center 200
"Isosurface Stuffing: Fast Tetrahedral Meshes with Good
Dihedral Angles"
Jonathan Shewchuk
Computer Science, UC Berkeley
http://graphics.stanford.edu/courses/cs528/
Abstract below
TUESDAY, 10 APRIL 2007
12 noon Berkeley On the Future of Scholarly Communication Seminar [10-Apr-07]
CSHE Library, South Hall Annex (Berkeley)
"DLF Aquifer: Improving Access to Distributed Scholarly Repositories"
Katherine Kott
Digital Library Federations: Aquifer Program
http://cshe.berkeley.edu/events/
Abstract below
12:30pm Berkeley Redwood Center for Theoretical Neuroscience [10-Apr-07]
3105 Tolman (Berkeley)
"Unsupervised discovery of structure for transfer learning"
Andrew Ng
Stanford University
http://redwood.berkeley.edu/seminars.php
3:30pm Stanford Semantics and Pragmatics Workshop [10-Apr-07]
Margaret Jacks Hall 460:126
"Incomplete Events, Incomplete Objects"
Zoltan Szabo
Yale University
http://www-linguistics.stanford.edu/semgroup/
Abstract below
4:10pm Berkeley Tanner Lecture on Human Values I [10-Apr-07]
Toll Room, Alumni House (Berkeley)
"On Public Reason"
Joshua Cohen
Political Science, Philosophy, Law, Stanford
http://www.grad.berkeley.edu/tanner/0607.shtml
6:30pm SULUG Meeting [10-Apr-07]
Gates 104
"on Numenta"
Bobby Jaros
Numenta
http://www.numenta.com/about-numenta/numenta-technology.php
http://sulug.stanford.edu/
7:30pm BayCHI [10-Apr-07]
George Pake Auditorium at PARC
"Designing for Lifestyle"
Kelly Goto
"Learning Interaction Design from Las Vegas"
Dan Saffer
http://www.baychi.org/program/
Abstract below
WEDNESDAY, 11 APRIL 2007
10:00am UC Berkeley CITRIS Distinguished Speaker [11-Apr-07]
290 Hearst Memorial Mining Building (UC Berkeley)
"Towards Personal, Proactive, & Preventive Healthcare:
Disruptive Technologies for the Global Age Wave"
Eric Dishman
Intel
http://www.citris-uc.org/
Abstract below
3:45pm Psychology Department Colloquium [11-Apr-07]
Jordan Hall 420:041
"The Mirror Neuron System in Monkeys and Humans"
Giacomo Rizzolatti
Universita di Parma
http://www-psych.stanford.edu/events_colloquium.html
4:10pm UC Berkeley 19th Annual Alfred Tarski Lecture 2 [11-Apr-07]
277 Cory Hall (Berkeley)
"Interpretations of Set Theory in Discrete Mathematics and
Informal Thinking: Interpreting Set Theory in Discrete
Mathematics: Boolean Relation Theory"
Harvey M. Friedman
Mathematics, Computer Science, and Philosophy, Ohio State University
http://logic.berkeley.edu/tarski-lectures.html
4:10pm Berkeley Tanner Lecture on Human Values II [11-Apr-07]
Toll Room, Alumni House (Berkeley)
"Democracy's Public Reason, Global Public Reason"
Joshua Cohen
Political Science, Philosophy, Law, Stanford
http://www.grad.berkeley.edu/tanner/0607.shtml
4:15pm EE380: Computer Systems Lab Colloquium [11-Apr-07]
Gates B01 (HP Auditorium)
Title to be announced
Paul Saffo
Institute for the Future
http://ee380.stanford.edu/contents.html
6:00pm Berkeley History and Philosophy of Logic Mathematics, and Science
234 Moses (Berkeley) [11-Apr-07]
"Empirical Investigations of Explanation and Causation"
Tania Lombrozo
Psychology, Berkeley
http://hplms.berkeley.edu/
Abstract below
6:30pm SF Bay ACM Data Mining SIG [11-Apr-07]
SAP LABS, Building D, 3410 Hillview Avenue, Palo Alto, CA
"Using Data Mining to Measure Similarity Between Words and Objects"
Mehran Sahami
Senior Research Scientist, Google
http://sfbayacm.org/dmsig.php
Abstract below
6:00pm Silicon Valley Web Guild [11-Apr-07]
Google, 1600 Amphitheatre Parkway, Mountain View
"Usability 2.0"
Panel
http://www.webguild.org/
Abstract below
THURSDAY, 12 APRIL 2007
4:00pm PARC Forum [12-Apr-07]
George Pake Auditorium at PARC
"Thinking big: Carbon capture and sequestration as a global
approach to greenhouse gas emission reduction"
S. Julio Friedmann
Lawrence Livermore National Laboratory
http://www.parc.com/forum/
4:00pm UC Berkeley CIS Seminar [12-Apr-07]
Soda Hall 310 (UC Berkeley)
"Oracle inequalities in sparse learning problems"
Vladimir Koltchinskii
School of Mathematics, Georgia Tech
http://www.cs.berkeley.edu/~feisha/cis-seminar
Abstract below
4:00pm UC Berkeley Oxyopia Lecture [12-Apr-07]
489 Minor Hall (UC Berkeley)
"What Are Illusions and Why Do We See Them?"
Beau Lotto
University College, London
http://optometry.berkeley.edu/opt_txtpp/ce/oxyopias.html
Abstract below
4:10pm Berkeley Tanner Lecture on Human Values: Seminar [12-Apr-07]
Toll Room, Alumni House (Berkeley)
Joshua Cohen
Political Science, Philosophy, Law, Stanford
http://www.grad.berkeley.edu/tanner/0607.shtml
4:15pm SSP10: Symbolic Systems Forum [12-Apr-07]
Bldg. 380:380C (Math Corner)
"Blocking and the System of Grammar"
Peter Sells
Linguistics, Stanford
http://symsys.stanford.edu/ssp_events
Abstract below
FRIDAY, 13 APRIL 2007
12:30pm CS547: Human-Computer Interaction Seminar [13-Apr-07]
Gates B03 (NEC Auditorium)
"Gaze-Enhanced User Interface Design"
Manu Kumar
Stanford HCI Group
http://hci.stanford.edu/cs547/
Abstract below
3:15pm Philosophy Department Colloquium [13-Apr-07]
Bldg. 90:92Q
"Influxus intelligibilis: Oughts, Requirements and Kant's
Concept of Practical Reason"
Konstantin Pollok
Stanford Humanities Center
http://www-philosophy.stanford.edu/ce.html
3:15pm Friday Cognitive Seminar [13-Apr-07]
Jordan Hall 420:050
"Ripples in the brain: Analyses of fMRI response patterns,
retinotopic position and object category"
Rory Sayres
http://www-psych.stanford.edu/events_cognitive.html
3:30pm Linguistics Department Colloquium [13-Apr-07]
Margaret Jacks Hall 460:126
title to be announced
Anthony Kroch
Univ. of Pennsylvania
http://www-linguistics.stanford.edu/colloq/
4:00pm UC Berkeley Oxyopia Lecture [13-Apr-07]
489 Minor Hall (UC Berkeley)
"New Ways of Analyzing the Functional Architecture of Human
Information Processing"
Charles Chubb
Cognitive Sciences, UC Irvine
http://optometry.berkeley.edu/opt_txtpp/ce/oxyopias.html
Abstract below
4:10pm UC Berkeley 19th Annual Alfred Tarski Lecture 2 [13-Apr-07]
60 Evans Hall (Berkeley)
"Interpretations of Set Theory in Discrete Mathematics and
Informal Thinking: Interpreting Set Theory in Informal
Thinking: Concept Calculus"
Harvey M. Friedman
Mathematics, Computer Science, and Philosophy, Ohio State University
http://logic.berkeley.edu/tarski-lectures.html
4:15pm CSLI Seminar on Computational Learning and Adaptation [13-Apr-07]
Cordura Hall 100
"From Transient Patterns to Persistent Relational Structures
Episodic Memory formation via Cortico-hippocampal Interactions"
Lokendra Shastri
International Computer Science Institute, Berkeley
http://cll.stanford.edu/scla/schedule.shtml
Abstract below
SUNDAY, 15 APRIL 2007
all day Stanford Community Day [15-Apr-07]
Stanford Campus
http://communityday.stanford.edu/
____________
Stanford Blood Center: Shortage of O-, A-, and B-. 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.
____________
STANFORD CHALLENGE ON K-12 EDUCATION TALK
on Wednesday, 4 April 2007, 5:00pm - 6:30pm
Cubberley Auditorium
http://ed.stanford.edu/
"Teaching Science: How, What, and Who Decides?"
Donald Kennedy
Editor-in-Chief of Science
Many of us know that students are losing interest in science education
as they grow older and move farther along in the K-12 educational
system. But there is little agreement about what is needed to improve
science education in the United States. Should children learn how to
actively engage in science through ways that encourage learning the
"whats" and "hows" of the scientific process? Or instead, should
teachers go "back to basics" with a more traditional approach that
emphasizes vocabulary and memorizing scientific facts? Don Kennedy,
Stanford President Emeritus and editor-in-chief of Science, will
explore how scientists are thinking about K-12 science teaching -- and
why scientific insights about how the human brain might connect to
learning have been disappointing. Part of the discussion will focus on
the critics of science teaching, with some attention to the increasing
efforts by Christian conservatives to challenge the teaching of
evolution in public schools.
This event kicks off The Stanford Challenge's initiative on Improving
K-12 Education, a campus-wide effort that brings together scholars
from across the university to identify and demonstrate strategies for
making fundamental improvements in the way our nation educates its
children.
A Q&A session and reception will immediately follow the lecture.
About the Speaker: Don Kennedy has served on the faculty of Stanford
University since 1960. He is presently a Senior Fellow at the Center
for Environmental Sciences & Policy. From 1980 to 1992, he served as
the President of Stanford University. Prior to his presidency, Kennedy
was the Director of the Program in Human Biology and the Chair of the
Biology Department. Apart from Stanford, Kennedy was Commissioner of
the U.S. Food & Drug Administration from 1977-79 and is a member of
the National Academy of Sciences, the American Academy of Arts and
Sciences, and the American Philosophical Society. He serves as a
director of the Carnegie Endowment and co-chair of the National
Academies' Project on Science, Technology and Law. Kennedy received AB
and PhD degrees in biology from Harvard University.
____________
SSP10: SYMBOLIC SYSTEMS FORUM
on Thursday, 5 April 2007, 4:15pm
Bldg. 380:380C (Math Corner)
http://symsys.stanford.edu/ssp_events
"Reward, Value and Choice:
A Perspective on the Neurobiology of Decision Making"
Bill Newsome
Neurobiology, Stanford
http://monkeybiz.stanford.edu/
Mammals have evolved highly sophisticated mechanisms for efficient
harvesting of rewards in an uncertain environment. An animal's recent
history of choices and rewards permits near-optimal estimates of the
"value" of a particular choice or action in terms of the probability
of acquiring an associated reward. To study the neural mechanisms
underlying this behavior, we trained rhesus monkeys on a version of
Hernnstein's classic "matching" task. Quantitative analysis of the
behavior permits a precise characterization of the computation that
the monkeys use to estimate reward probability, and neurophysiological
recordings have revealed potential neural substrates. We are currently
engaged in neuroimaging studies (fMRI) to identify additional brain
regions that are likely to contribute to value computations. The
current studies are central to an emerging neurobiology of decision
making.
____________
BERKELEY INSTITUTE OF COGNITIVE AND BRAIN SEMINAR
on Friday, 6 April 2007, 11:00am
Tolman 5101 (Berkeley)
http://icbs.berkeley.edu/
"Differentiation and Disintegration of Conceptual Knowledge:
A Parallel-Distributed Processing Approach"
Jay McClelland
Psychology, Stanford
I will describe a model of human semantic cognition, based on the
ideas of distributed representation and gradual incremental learning
inherent in the Parallel-Distributed Processing (PDP) framework. The
model addresses progressive differentiation of conceptual knowledge in
development and progressive disintegration of conceptual knowledge in
semantic dementia, a rare condition affecting the anterior and lateral
portions of the temporal lobes. In addition, the model addresses
primacy of basic level naming responses in development and dementia,
as well as typicality, frequency, and expertise effects, providing a
broad framework for understanding development, adult cognitive
performance, and deterioration of conceptual knowledge in dementia.
The model also addresses many of the phenomena that some have taken as
supporting the idea that human semantic knowledge takes the form of
naive, implicit domain theories, including category coherence effects,
differential importance of different properties for different domains,
and reorganization of conceptual knowledge in development. It suggests
how domain specific constraints on the interpretation of new
information may arise from prior experience, providing an alternative
to nativist approaches to the origins of such constraints on
cognition.
____________
CS547: HUMAN-COMPUTER INTERACTION SEMINAR
on Friday, 6 April 2007, 12:30-2:00pm
Gates B03 (NEC Auditorium)
http://hci.stanford.edu/cs547/
"Paying Attention to Interruption:
A Human-Centered Approach to Intelligent Interruption Management"
Brian Bailey
University of Illinois (UIUC)
http://www-faculty.cs.uiuc.edu/%7Ebpbailey/
Proactive computing offers many desired benefits to users, such as
enabling a high degree of awareness of peripheral information.
However, notifications from proactive systems run the serious risk of
interrupting users' tasks at inopportune moments, decreasing
performance and increasing frustration. In this talk, I will discuss
our ongoing empirical and systems development work aimed at
maintaining timely delivery of notifications while reducing costs of
interruption. Grounded in cognitive theories of attention, a central
theme of our work has been to leverage deeper understanding of the
structure of user tasks in order to pinpoint lower cost moments for
interruption. Our empirical work has shown favorable results for
deferring notifications until meaningful breakpoints in a task, while
our systems development work has begun to understand how to
efficiently automate this process in practice. The work discussed
should be applicable in most situations in which users perform
goal-directed tasks in the midst of periodic interruption.
About the Speaker: Brian Bailey has been an Assistant Professor in the
Department of Computer Science at the University of Illinois-Urbana
since 2002. His research interests include developing
theoretically-based tools, interfaces, and systems for managing user
attention, designing computational tools that better support
creativity and storytelling, and user interfaces for pervasive
computing. Dr. Bailey's multi-disciplinary research efforts have been
recognized with affiliate academic appointments in Aviation
Psychology, the Beckman Institute, and the Graduate School of Library
and Information Science at UIUC.
____________
BERKELEY INTERNATIONAL COMPUTER SCIENCE INSTITUTE
on Friday, 6 April 2007, 1:30pm
Main Lecture Hall, ICSI, 1947 Center Street, Sixth Floor, Berkeley
http://www.icsi.berkeley.edu/
"Ensembles of classifiers"
Alberto Suarez
Universidad Autonoma de Madrid and ICSI
Numerous experimental studies show that pooling the decisions of
classifiers in an ensemble can lead to improvements in the
generalization performance of weak learners. In the first part of this
talk we present two novel algorithms to generate ensembles of
classifiers. The first method takes advantage of the intrinsic
variability of the iterative growing and pruning tree construction
procedure devised by Gelfand et al. [Gelfand, S.B., Ravishankar, C.S.
and Delp, E.J. (1991), "An Iterative Growing and Pruning Algorithm for
Classification Tree Design", IEEE Transactions On Pattern Analysis and
Machine Intelligence Vol. 13, no. 2 138-150].
The second procedure consists in generating classifiers of the same
type built using perturbed versions of the training data. These
perturbed data are obtained by injecting random uncorrelated noise in
the class labels of the original training examples. The final
classification is obtained by averaging over classifiers obtained with
independent realizations of the noise.
The second part of the talk is devoted to exploring the importance of
the order of aggregation in bagging ensembles. We present several
heuristic rules to order the classifiers in bagging ensembles prior to
aggregation.
Early stopping in the aggregation of the classifiers ordered with
these rules allow the identification of subensembles that require less
memory for storage, have a faster classification speed and can perform
better than the original bagging ensemble. Furthermore, this ensemble
pruning procedure does not seem to deteriorate the robust performance
of bagging in noisy classification tasks.
About the Speaker: Alberto Suarez received the degree of Licenciado in
Chemistry from the Universidad Autonoma de Madrid, Madrid, Spain, in
1988, and the Ph.D. degree in Physical Chemistry from the
Massachusetts Institute of Technology (MIT), Cambridge, MA, in
1993. After holding postdoctoral positions at Stanford University
(USA), at the Universite Libre de Bruxelles (Belgium), as a research
fellow financed by the European Commission within the program
"Training and Mobility of Researchers", and at the Katholieke
Universiteit Leuven (Belgium), he is currently a professor in the
Computer Science Department of the Universidad Autonoma de Madrid
(Spain). He has worked on relaxation theory in condensed media,
stochastic and thermodynamic theories of nonequilibrium systems,
lattice-gas automata, and decision tree induction. His current
research interests include machine learning, quantitative and
computational finance, time series analysis and information processing
in the presence of noise.
____________
BERKELEY INFORMATION ACCESS SEMINAR
on Friday, 6 April 2007, 3:00pm - 5:00pm
107 South Hall (Berkeley)
http://courses.ischool.berkeley.edu/i296a-1/s07/schedule.html
"Objects with Narrative Memories"
Kimiko Ryokai
School of Information, Berkeley
My work builds on the relationships people have with their physical
objects. Physical objects are charged with history, narratives, and
memories of people, both from the ones who created them and from the
ones who interacted with them. However, we usually do not have access
to this information about our physical objects. With my past and
present projects, StoryMat and I/O Brush, I will discuss the potential
that technologies have in creating that missing link.
About the Speaker: Kimiko Ryokai is our School's newest faculty
member. She is also a faculty member in the Center for New Media. See
her profile at
http://www.ischool.berkeley.edu/people/faculty/kimikoryokai/
____________
LINGUISTICS DEPARTMENT COLLOQUIUM
on Friday, 6 April 2007, 3:30pm
Margaret Jacks Hall 460:126
http://www-linguistics.stanford.edu/colloq/
"Tattered Syntaxes: Beckett's Late Style"
Ann Banfield
UC Berkeley
Samuel Beckett's oeuvre remains one of the last productions of
literary modernism, resisting all attempts at recuperation by its
recalcitrant strangeness. It provides a counter-example to the
received idea that each successive style that initially puzzles by its
difficulty finally ceases to shock. Beckett's work was preceded by one
presenting an equal though not the same strangeness: Finnegans Wake.
The difference between Joyce and himself, Beckett once explained, is
that Joyce tended "toward omniscience" whereas he worked with
"ignorance". The language of Finnegans Wake is the language of a
knower; Beckett's minimalist language is that of a "non-knower, a
non-can-er", in a quite precise linguistic sense. Beckett and Joyce
each exploited a different component of the grammar, specifically of
the lexicon, the one aiming at maximal production and in the other at
minimal. Joyce's route was lexical and Beckett's syntactic. Hugh
Kenner makes the distinction between "a word-man or a sentence-man".
Taking "word" in its literal sense as the productive, lexical word,
Joyce's was a "Revolution of the word," as Eugene Jolas put it.
"Joyce's repertory of syntactic devices is not extensive", Kenner
observes. "He is not, like Beckett, an Eiffel nor a Calder of the
sentence. The single word--'repaired'; 'salubrious'--is his normal
means to his characteristic effects." Kenner's insight only goes so
far. It is not that Beckett was an architect of the sentence. Beckett
called his the "literature of the unword." He spoke of "tattered
syntaxes," "scraps of an ancient voice in me not mine". Fragments of
syntax, but well-formed fragments, syntactic structures--phrasal
categories, maximal projections.
A more precise formulation of Kenner's distinction is given by X-Bar
Syntax, together with Emonds' version of the division between lexical
words and grammatical formatives: Dictionary and "Syntacticon." The
latter includes not just functional categories and bound morphemes but
a non-productive set of grammatical nouns, verbs and adjectives.
Syntacticon items map into the skeleton of X-Bar Syntax, where the
basic and largest syntactic units (the highest, maximal projections)
are phrasal categories in general., of which the sentence is only one
actualization. Beckett intuits this leveling of the sentence to the
structure of all major phrasal categories as XP. Wakese uses the
generative processes for the production of new lexical items.
Phonology, morphology and etymology underlie its lexical
inventiveness. Emonds' Dictionary is the interface with non-linguistic
memory and culture; its members have full semantic features. Hence
Joyce's "omniscience". The young Beckett imitated but ultimately
resisted Joyce's productivity as a gift inherited against his
will--"what a gazeteer I am", the Unnamable jests.
Beckett's late style is created largely out of the Syntacticon, while
scarcely exploiting the Dictionary. Because it is language's relation
to Kant's synthetic a priori, the Syntacticon is the source for the
philosophical vocabulary. This explains the perceived philosophical
quality of Beckett's work. Its abstractness lies in the fact that
Syntacticon items have only syntactic or cognitive features. The
"philosophy" that emerges from it is grounded in Beckett's
exploitation of syntacticon items such as quantifiers, especially the
comparative, and directional prepositions, especially the intransitive
Ps he brought to the fore in his returns to English, the philosophy of
a directionality or "path" without movement, heading in a lesswards or
worstward direction: "stirring still" ever less and less but always
just short of nothing.
____________
SRI CSL SEMINAR SERIES
on Friday, 6 April 2007, 3:30pm - 4:30pm
EK255, SRI International
http://www.csl.sri.com/
"Overview of Real-Time Maude and its Applications"
Peter Olveczky
University of Oslo
Real-Time Maude is a tool that extends the rewriting logic-based Maude
system to support the executable formal modeling and analysis of
real-time systems. Real-Time Maude is characterized by its general and
expressive, yet intuitive, specification formalism, and offers a
spectrum of formal analysis methods, including: rewriting for
simulation purposes, search for reachability analysis, and temporal
logic model checking. Our tool is particularly suitable to specify
real-time systems in an object-oriented style, and its flexible
formalism makes it easy to model different forms of communication.
This modeling flexibility, and the usefulness of Real-Time Maude for
both simulation and model checking, has been demonstrated in advanced
state-of-the-art applications, including scheduling and wireless
sensor network algorithms, and communication and cryptographic
protocols.
This talk gives a high-level overview of Real-Time Maude and some of
its applications, and briefly discusses completeness of analysis for
dense-time systems.
____________
CSLI SEMINAR ON COMPUTATIONAL LEARNING AND ADAPTATION
on Friday, 6 April 2007, 4:15pm to 5:30pm
Cordura Hall 100
http://cll.stanford.edu/scla/schedule.shtml
"Distributed Detection and Inference:
An adaptive anomaly detector for worm detection"
John Mark Agosta
Machine Learning Group, Intel Research
We present a worm traffic detector that monitors an end-host's
out-going traffic, in a way that is suitable for integration into a
larger distributed intrusion detection system. Conventional
centralized systems observe traffic at central points and have limited
visibility into network traffic and host machine state, as well as
limited resources that they can bring to detection. A distributed
system that incorporates the end-host detector that we present is not
constrained by these limitations, and additionally can accommodate
variations across machine behavior. We explore the idea of such
adaptive end-host detectors, where a classifier trained as a traffic
predictor is used to customize the host's worm detection threshold as
a function of time. Using real traffic traces collected from a number
of end-hosts and superimposing a worm against this traffic, we show
that a rudimentary predictor and threshold adaptation model
considerably improve anomaly detection. We show that measures of the
traffic predictor's error rate, the reduction in the "threshold gap,"
and the ability to detect the simulated threat strongly agree.
Based on joint work with Carlos Diuk-Wasser, CS Department,
Rutgers University, and Jaideep Chandrashekar and Carl Livadas, Intel
Research.
____________
CS528: BROAD AREA COLLOQUIUM
on Monday, 9 April 2007, 4:15pm - 5:30pm
Hewlett Teaching Center 200
http://graphics.stanford.edu/courses/cs528/
"Isosurface Stuffing:
Fast Tetrahedral Meshes with Good Dihedral Angles"
Jonathan Shewchuk
Computer Science, UC Berkeley
The isosurface stuffing algorithm fills an isosurface with a uniformly
sized tetrahedral mesh whose dihedral angles are bounded between 10.7
degrees and 165 degrees. All vertices on the boundary of the mesh lie
on the isosurface. The algorithm is whip fast, numerically robust, and
easy to implement because, like Marching Cubes, it generates
tetrahedra from a small set of precomputed stencils. A variant of the
algorithm creates a mesh with internal grading: on the boundary, where
high resolution is generally desired, the elements are fine and
uniformly sized, and in the interior they may be coarser and vary in
size. Isosurface stuffing is the first algorithm that simultaneously
copes with boundaries of complex shape and rigorously guarantees the
suitability of its tetrahedra for finite element methods. Our angle
bounds are guaranteed by a computer-assisted proof. We illustrate the
use of isosurface stuffing for dynamic remeshing in a fluid simulation
with moving liquid surfaces.
About the Speaker: Jonathan Shewchuk is an Associate Professor in the
Department of Electrical Engineering and Computer Sciences at UC
Berkeley. He is best known for his Triangle software for high-quality
triangular mesh generation, and his "Introduction to the Conjugate
Gradient Method Without the Agonizing Pain".
____________
BERKELEY ON THE FUTURE OF SCHOLARLY COMMUNICATION SEMINAR
on Tuesday, 10 April 2007, 12 noon
CSHE Library, South Hall Annex (Berkeley)
http://cshe.berkeley.edu/events/
"DLF Aquifer: Improving Access to Distributed Scholarly Repositories"
Katherine Kott
Digital Library Federations: Aquifer Program
The Digital Library Federation seeks to provide scholars, particularly
social and cultural historians, literary scholars and scholars working
in interdisciplinary fields, with consistent access to digital library
collections pertaining to nineteenth and twentieth century United
States social history across numerous institutional boundaries through
it's DLF Aquifer initiative. The project is designed to address the
difficulty humanities and social science scholars face in finding and
using digital materials that are often located in a variety of
environments with a bewildering array of interfaces, access protocols
and usage requirements. By gathering distributed collections and
integrating them into a variety of local environments, the project
will bring the resources to the scholar and make material from remote
repositories available through locally supported tools. This
presentation will describe the initiative, which is in the project
planning stage and solicit suggestions from participants for
successful implementation.
About the Speaker: Katherine Kott is the director of the Digital
Library Federation's DLF Aquifer program. Her professional career has
included a wide range of responsibilities in libraries and information
services. Prior to beginning her work with the Digital Library
Federation in 2005, Kott was the head of cataloging and metadata
services at Stanford University, where she is based. Before arriving
at Stanford, she led the implementation services department at
Innovative Interfaces, Inc., managing the installation of integrated
library systems around the world. She has promoted the idea of
leveraging resources through collaboration throughout her career,
including work as a systems librarian at Bates College and in law
library technical services at Duke University.
____________
STANFORD SEMANTICS AND PRAGMATICS WORKSHOP
on Tuesday, 10 April 2007, 5:30pm
Margaret Jacks Hall 460:126
http://www-linguistics.stanford.edu/semgroup/
"Incomplete Events, Incomplete Objects"
Zoltan Szabo
Yale University
This paper discusses the semantics of the progressive aspect and defends
a substantially revised version of Terence Parsons's approach. The
revisions come in response to two main charges. The first is that
Parsons's account does not explain why `John crossed the street' entails
`John was crossing the street' (and in general, why perfective
accomplishments entail their progressive counterparts). The second is
that Parsons's account incorrectly predicts that `John is building a
house' entails `There is a house John is building' (and in general, that
the direct object of progressive accomplishments is available for
existential generalization).
____________
BAYCHI
on Tuesday, 10 April 2007, 7:30pm - 9:30pm
George Pake Auditorium, PARC
http://www.baychi.org/program/
"Designing for Lifestyle"
Kelly Goto
Interaction design is no longer limited to the web. The concept of
user experience is being redefined as multiple delivery methods of
social and business interaction merge into our lifestyles. As design
migrates from the web to mobile devices we carry and interact with on
a daily basis, our approach must also shift into cycles of design and
research centered around the way people actually live.
In an enlightening session, Kelly Goto will discuss the evolution of
web, handheld, and product interfaces and their cultural impact. Learn
how companies are utilizing ethnographic-based research to conduct
rapid, immersive studies of people and their lifestyles to inform the
usefulness and viability of interfaces both online and offline.
About the Speaker: Kelly Goto BioAs an evangelist for "design
ethnography", Kelly Goto is dedicated to understanding how real people
integrate products and services into their daily lives. Goto is
principal of gotomedia, LLC, a global leader in research-driven,
people-friendly interface design for web, mobile and product solutions
for clients including Seiko Epson Japan, Adobe, NetIQ, WebEx and
CNET. Her book, Web Redesign 2.0: Workflow That Works, is a standard
for user-centered design principles. Goto is also the editor of
gotomobile.com, a leading online publication on mobile user experience
and serves on the national board of the AIGA Center for Brand
Experience.
"Learning Interaction Design from Las Vegas"
Dan Saffer
The hit of SXSW replayed! MySpace is junk. MySpace is ugly. You hear
designers say this often. Funny, architects thought the same thing
about Las Vegas until the book Learning from Las Vegas revealed its
hidden patterns. Las Vegas has much to teach about design--and about
MySpace--for those who can see why it is as it is. Find out how you
can apply the patterns of Las Vegas to your website or device.
About the Speaker: Dan Saffer is an interaction designer at Adaptive
Path and the author of Designing for Interaction: Creating Smart
Applications and Clever Devices (New Riders 2006). His work and
writings on design have appeared in such publications as Entertainment
Weekly and BusinessWeek. He has a masters degree in interaction design
from Carnegie Mellon, and curates a site that collects physical
interfaces called No Ideas But In Things.
____________
UC BERKELEY CITRIS DISTINGUISHED SPEAKER
on Wednesday, 11 April 2007, 10:00am
290 Hearst Memorial Mining Building (UC Berkeley)
http://www.citris-uc.org/
"Towards Personal, Proactive, & Preventive Healthcare:
Disruptive Technologies for the Global Age Wave"
Eric Dishman
Intel
In the midst of disruptive demographic changes that will double the
number of people over age 60 to more than 1.2 billion by 2025, we are
faced with an enormous global challenge: how do we improve the quality
of life for double or even triple the number of people in need while
decreasing the costs of healthcare? The answer is most certainly not
to continue the healthcare "business as usual" approach we take today
which is premised upon reaction-to-crisis, hospital settings, and
late-in-the-game intervention. It is imperative that we transform
healthcare from reactive to proactive, from population-based to
personal, from intervention to prevention, and from hospital to home.
Since 1999, Intel social scientists have been studying the needs of
people across the continuum of care to figure out how technologies
might be used to drive a more proactive paradigm. Starting in 2005,
Intel launched its newest business, the Digital Health Group, to begin
developing platforms for proactive healthcare. In this talk, Eric
Dishman, General Manager for Intel's Health Research & Innovation
Group ( http://www.intel.com/healthcare/hri/ ), will give an overview
of Intel's strategy and research to promote prevention, early
detection, compliance, and family caregiving through disruptive
technologies such as wireless sensor networks, Web 2.0 capabilities,
cell phones, WiMax, behavioral analytic engines, and more. In
particular, he will focus on Intel's worldwide efforts to accelerate
R&D on independent living and chronic disease management technologies
that will help older citizens to better care from themselves from
wherever they call home. His talk will include Intel's policy efforts
in CAST ( http://www.agingtech.org/ ) and the new TRIL Centre (
http://www.trilcentre.org/ ) in Ireland; several prototype
technologies will be shown in the areas of Alzheimer's assistance,
Parkinson's treatment, medication compliance, social health, and the
prevention of cardiovascular disease.
____________
BERKELEY HISTORY AND PHILOSOPHY OF LOGIC, MATHEMATICS, AND SCIENCE
on Wednesday, 11 April 2007, 6:00pm - 7:30pm
234 Moses (Berkeley)
http://hplms.berkeley.edu/
"Empirical Investigations of Explanation and Causation"
Tania Lombrozo
Psychology, Berkeley
What constitutes an explanation? This question has received
considerable attention in philosophy of science, but relatively little
is known about the psychology of explanation. In this talk I'll
present recent empirical work on the psychology of explanation,
focusing on teleological explanations--explanations in terms of a
function or goal. Drawing on analyses from philosophy, I'll suggest
that people understand teleological explanations as causal
explanations: they are only accepted when the function invoked in the
explanation played a causal role in bringing about what is being
explained. However, I'll also suggest that casual relationships are
evaluated differently in the context of teleological explanations than
in the context of mechanistic explanations. These differences map onto
different philosophical theories of causation-- roughly, a
counterfactual account versus a physical connection/transfer of force
account. Thus a second, more speculative aim of the talk will be to
suggest the psychological reality of multiple concepts of causation.
____________
SF BAY ACM DATA MINING SIG
on Wednesday, 11 April 2007, 6:30pm - 9:00pm
SAP LABS, Building D, 3410 Hillview Avenue, Palo Alto, CA
http://sfbayacm.org/dmsig.php
"Using Data Mining to Measure Similarity Between Words and Objects"
Mehran Sahami
Senior Research Scientist, Google
The World Wide Web provides a wealth of data that can be harnessed to
help improve information retrieval and increase understanding of the
relationships between different entities. In many cases, we are often
interested in determining how similar two entities may be to each
other, where the entities may be pieces of text or descriptions of
some object. In this work, we examine multiple instances of this
problem, and show how they can be addressed by harnessing data mining
techniques applied to large web-based data sets. Specifically, we
examine the problems of determining the similarity of short texts
(even those that may not share any terms in common) and also of
learning similarity functions for semi-structured data to address
tasks such as record linkage between objects. While we present rather
different techniques for each problem, we show how measuring
similarity between entities in these domains has a direct application
to the overarching goal of improving information access for users of
web-based systems.
About the Speaker: Mehran Sahami is a Senior Research Scientist at
Google. His research interests include machine learning, data mining,
and information retrieval on the Web. Mehran was also previously a
Lecturer in the Computer Science Department at Stanford University
(where he received his PhD), and prior to Google, involved in a number
of commercial and research machine learning projects at Epiphany,
Xerox PARC and Microsoft Research. He has published dozens of refereed
technical papers, served on numerous conference program/organizing
committees and has several patents pending. This year he is serving at
Track Chair for the Industrial Practice and Experience track at WWW-07
and is Co-Chair of the Student Abstract and Poster program at AAAI-07.
____________
SILICON VALLEY WEB GUILD
on Wednesday, 11 April 2007, 6:00pm
Google, 1600 Amphitheatre Parkway, Mountain View
http://www.webguild.org/
(registration and payment)
"Usability 2.0"
Sean Kane, Director, User Interface Engineering, Netflix
Jon Wiley, User Experience Designer, Google
and a third to be announced
Google Apps, Netflix What is the next generation of web usability?
What are some of the usability gains and challenges arising as a
result of Web 2.0 technologies such as Ajax; how is usability viewed
within companies today; and what is the ROI of good usability
practice. Learn about the next generation or evolution of usability
from usability 1.0 to 2.0 and beyond on the web including what has
changed, and what are the challenges and the implications for
designers, developers, users, and marketers. The panel will also speak
to top usability trends, best practices, usability dos & don'ts,
techniques, processes, quantitative versus qualitative considerations,
etc. to creating highly usable and successful sites and webapps. Learn
from the pros in charge of highly successful sites how to make your
webapps and sites highly usable and successful!
____________
UC BERKELEY CIS SEMINAR
on Thursday, 12 April 2007, 4:00pm-5:00pm
Soda Hall 310 (UC Berkeley)
http://www.cs.berkeley.edu/~feisha/cis-seminar
"Oracle inequalities in sparse learning problems"
Vladimir Koltchinskii
School of Mathematics, Georgia Tech
A number of problems in Statistics and Learning Theory, such as
regression and classification, can be formulated as risk minimization
over a linear span of a large dictionary of given functions. Several
methods have been developed to find a "sparse" solution of such a
problem (provided that it exists). One of these methods is penalized
empirical risk minimization with $\ell_1$-norm of the vector of
coefficients used as a penalty. Another method is so called "Dantzig
Selector" recently introduced by Candes and Tao. We will discuss
several oracle inequalities that express "adaptivity" of these methods
to unknown sparsity of the problem.
____________
UC BERKELEY OXYOPIA LECTURE
on Thursday, 12 April 2007, 4:00pm
489 Minor Hall (UC Berkeley)
http://optometry.berkeley.edu/opt_txtpp/ce/oxyopias.html
"What Are Illusions and Why Do We See Them?"
Beau Lotto
University College, London
Seeing colour is arguably the most basic of visual attributes. And
yet, human perception of colour remains enigmatic largely because we
frequently see the world "incorrectly" -- in other words, because we
see illusions. Thus, illusions are a key tool for investigating how
and why we see what we do. In this talk, I will present our recent
work on human, bumblebee and synthetic system vision, which provides
evidence for the burgeoning hypothesis that illusions are caused by
(i) the ambiguity of visual stimuli, and (ii) their empirical -- and
thus statistical -- resolution. These studies (and this framework
generally) also suggest a more formal, quantitative definition of
illusion, as well as the basis for explaining the computational and
mechanistic principles that underlie what we (and bees) see.
____________
SSP10: SYMBOLIC SYSTEMS FORUM
on Thursday, 12 April 2007, 4:15pm
Bldg. 380:380C (Math Corner)
http://symsys.stanford.edu/ssp_events
"Blocking and the System of Grammar"
Peter Sells
Linguistics, Stanford
'Blocking' is the name often given to a well-known phenomenon where
one linguistic form appears to pre-empt the use of another. In
English, 'bigger' is often used in preference to 'more big', even
though both intuitively mean the same; as linguists, we say that the
synthetic form 'bigger' blocks the analytic form 'more big' in certain
contexts. With other words, the situation is different: the
conditions on '-er' suffixation a putative word like 'substantialer'
is a highly dispreferred form, and hence 'more substantial' would be
used in (almost) all grammatical contexts. In other languages, we
often find that a particular inflected synthetic form blocks the more
analytic form with absolute regularity, and I will mention a few cases
in my talk. Hence, blocking of this kind is about the relation
between the linguistic duty that a word performs relative to the duty
that a phrase performs, when both intuitively express the same
content. Recently, there has been some focus on this in the
theoretical linguistic literature -- what the conditions are which
allow a blocking relationship, what the nature of the blocking
relationship is, and what grammar must be like, to have those
properties. In the talk I will discuss some conclusions that can be
drawn about the nature of grammar, in terms of what architecture it
has and what kinds of mechanisms it has within that architecture.
____________
CS547: HUMAN-COMPUTER INTERACTION SEMINAR
on Friday, 13 April 2007, 12:30-2:00pm
Gates B03 (NEC Auditorium)
http://hci.stanford.edu/cs547/
"Gaze-Enhanced User Interface Design"
Manu Kumar
Stanford HCI Group
The eyes are a rich source of information for gathering context in our
everyday lives. Using eye-gaze information as a form of input can
enable a computer system to gain more contextual information about the
user's task, which in turn can be leveraged to design interfaces which
are more intuitive and intelligent. With the increasing accuracy and
decreasing cost of eye gaze tracking systems it will soon be practical
for able-bodied users to use gaze as a form of input in addition to
keyboard and mouse - provided the resulting interaction is an
improvement over current techniques. Our research explores how gaze
information can be effectively used as an augmented input in addition
to traditional input devices. We present a series of novel prototypes
that explore the use of gaze as an augmented input to perform everyday
computing tasks. In particular we explore the use of gaze-based input
for pointing and selection, application switching, password entry,
scrolling, zooming and document navigation. We present the results of
user experiments which compare the gaze-augmented interaction
techniques with traditional mechanisms and show that the resulting
interaction is either comparable to or an improvement over existing
input methods. These results show that it is indeed possible to devise
novel interaction techniques that use gaze as a form of input without
overloading the visual channel and minimizing false activations. We
also discuss some of the problems and challenges of using gaze
information as a form of input and proposes solutions which, as
discovered over the course of the research, can be used to mitigate
these issues.
For more information see the GUIDe Project page
http://hci.stanford.edu/research/GUIDe/
About the Speaker: Manu Kumar is a Doctoral Candidate in Computer
Science at Stanford University. Prior to joining Stanford, Manu
started two successful companies - SneakerLabs, Inc. which developed
web-based customer interaction tools and iMeet, Inc. which developed
webconferencing software. Manu served on the Board of Directors of
Netspoke, Inc. a company which provides both audio and web
conferencing. Manu holds a Bachelors degree in Electrical and Computer
Engineering with Honors and a Masters degree in Software Engineering
and Business, both from Carnegie Mellon University and a Masters in
Computer Science from Stanford University. He has worked on research,
development and commercialization of interactive technologies ranging
from chat, web conferencing, application sharing, distance learning,
audio conferencing, and CRM applications. He holds one patent in this
space with additional patents pending. His prior research has been in
the area of automobile interfaces.
____________
UC BERKELEY OXYOPIA LECTURE
on Friday, 13 April 2007, 4:00pm
489 Minor Hall (UC Berkeley)
http://optometry.berkeley.edu/opt_txtpp/ce/oxyopias.html
"New Ways of Analyzing the Functional Architecture of Human
Information Processing"
Charles Chubb
Cognitive Sciences, UC Irvine
Equisalience function analysis (EFA) is a new methodology for
investigating basic questions about the functional architecture of
human perception and cognition. EFA lets you test whether two
perceptual or cognitive processes have similar access to different
sorts of information in the sensory input stream. This talk gives an
introduction to the method and describes our research using it to
study the differential access of the ventral and dorsal visual
pathways to chromatic versus luminance information. Our results
suggest that three distinct visual processes (with different relative
sensitivities to luminance versus equiluminant chromatic variations)
are used for 1. shape discrimination, 2. localization of objects in
allocentric space, and 3. online control of aimed movements. The
research of others (e.g., Milner & Goodale, 1995; Creem & Proffitt,
2001; Glover & Dixon, 2002) suggests that these processes might
correspond to those hypothesized in the temporal lobe, the inferior
parietal lobule, and the superior parietal lobule, respectively. In
the last part of the talk, we discuss motion perception as potential
domain for EFA, presenting results from some preliminary experiments
to investigate the relative sensitivity of several motion-based
judgments to luminance versus texture-contrast.
____________
CSLI SEMINAR ON COMPUTATIONAL LEARNING AND ADAPTATION
on Friday, 13 April 2007, 4:15pm to 5:30pm
Cordura Hall 100
http://cll.stanford.edu/scla/schedule.shtml
"From Transient Patterns to Persistent Relational Structures
Episodic Memory formation via Cortico-hippocampal Interactions"
Lokendra Shastri
International Computer Science Institute, Berkeley
We readily remember events and situations in our daily lives and
acquire memories of specific facts by watching a telecast, reading a
newspaper, or participating in a dialogue. This one-shot mnemonic
ability poses a challenge for computational neuroscience: How does the
brain carry out this remarkable memorization task rapidly and
effortlessly?
There is a broad consensus that the hippocampal system (HS),
consisting of the hippocampal formation and neighboring cortical areas
plays a critical role in the encoding and retrieval of such "episodic"
memories. Furthermore, a great deal is known about the anatomy and
electrophysiology of the HS, including quantitative data about cell
counts and connectivity. But how the HS supports episodic memory
function is not well understood.
I will describe a circuit-level computational model (SMRITI*) that
demonstrates how a cortically expressed pattern of activity
representing an event can be transformed rapidly (~500 msec.) into a
robust memory trace by a neural structure analogous to the HS. The
model has a large memory capacity (> 50,000 events), displays a high
degree of pattern separation, and yet responds strongly to partial
cues.
The model suggests that there exists a striking correspondence between
the functional requirements of encoding episodic memories and the
idiosyncratic architecture and local circuitry of the HS. The model
also identifies constraints on the representation of relational
knowledge and predicts memory deficits that would result from insult
to specific HS subregions and cortical circuits projecting to the HS.
*System for Memorizing Relational Instances from Transient Impulses
____________
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). Problems with subscribing or
unsubscribing should be sent to
owner-csli-calendar@csli.stanford.edu