Old Email Archive

digest 1998-10-14 #001.txt


Tuesday
From: "Society for Literature & Science" 

Daily SLS Email Digest
-> AI and narrative symposium
by Ann Weinstone 
----------------------------------------------------------------------
Date: 13 Oct 1998 13:57:43 -0700
From: Ann Weinstone 
Subject: AI and narrative symposium
>Dear Colleagues,
>
>Michael Mateas and I are organizing a proposed 1999 AAAI Fall
Symposium
>on Narrative Intelligence.  We have noticed that in recent years
>narrative seems to be (re-)gaining popularity in various subfields
of AI
>and beyond.  We would like to get together people with interests in
>narrative stemming from various interdisciplinary excursions in
order to
>get a broad overview of what narrative does and could mean for AI.
>
>As part of our symposium proposal, we need to send in a list of
>potential participants.  A "potential participant" is
someone who is
>interested in attending the symposium; it does not imply any
obligation
>to actually attend.  If you think this symposium might be of
interest to
>you, please email me at phoebe@zkm.de with your name and
institutional
>affiliation.  Please feel free to forward this to those you think
might
>be interested.
>
>I have attached a (DRAFT) symposium proposal, which outlines our
>approach to NI and our goals for this symposium.  This should give
you a
>more concrete idea of what our symposium would be about.
>
>Cheers,
>Phoebe
>
>Phoebe Sengers
>Fulbright Scholar
>Center for Art and Media Technology (ZKM)
>Institute for Visual Media
>Postfach 6909
>D-76049 Karlsruhe
>Germany
>phoebe@zkm.deProposed Symposium for the AAAI 1999 Fall Symposium
Series:
>Narrative
>Intelligence
>
>
>While narrative has long been a theme in AI (see, for example,
[4A]),
>it has recently experienced a surge of popularity.  Researchers in
>various subfields, including story generation and understanding
[5],
>agent architecture [2] [9], and interface agents [7], have taken
>independent forays into narrative, finding it a fruitful way to
>rethink some basic issues in AI.  Our goal for our proposed
symposium,
>Narrative Intelligence (NI), is to bring researchers from these
>disparate perspectives together to talk about what we have learned
>about narrative and its application to AI.  In particular, we would
>like to explore the issues surrounding the construction of systems
---
>whether story systems, agents, or in other forms --- that produce
>behavior that humans can interpret as narrative.
>
>Some idea of the range of these approaches can be gleaned from the
>work of the members of the organizing committee, which is
>representative, but certainly not exhaustively so, of work in this
area:
>
>- Sengers [10] has studied how to build behavior-based believable
>agents that actively structure their behavior so as to be
narratively
>understandable by human observers. Borrowing ideas from Jerome
>Bruner's theory of narrative psychology [1], Sengers' system, the
>Expressivator, actively keeps track of what an agent has
communicated
>to the human participant. This knowledge is used by the agents to
>decide on transitions that maintain coherent narrative flow when
>switching between behaviors.
>
>- Mateas [3] is currently building Terminal Time, a system that
>constructs ideologically-biased documentary histories in response
to
>audience feedback.  Terminal Time pursues a set of rhetorical goals
in
>constructing its ideologically biased history. These goals are
>modified as the audience responds to questions intended to
determine
>the audience's own ideological bias. This project, a collaboration
>with documentary filmmaker Steffi Domike and interactive artist
Paul
>Vanouse, makes use of understandings of narrative in documentary
films
>and the arts.
>
>- Elliott [5] has built a system that, while using a fixed script,
>tells different stories by narrating the stories with different
>emotional emphases. This is an alternative to more traditional
>approaches to narrative generation which explore a combinatorial
space
>of primitive story elements. The emotional behavior of the
narration
>agent is generated by the Affective Reasoner, a cognitive appraisal
>model of emotion. Elliott's work demonstrates that human
understanding
>of narrative makes use of an understanding of emotion.
>
>- Dautenhahn and Nehaniv [2] have explored the use of narrative as
a
>means for behavior-based agents to represent themselves by
maintaining an
>autobiography. Narrative structure allows an agent to selectively
>foreground different aspects of its history.  Autobiographical
agents
>can share their histories with other agents by communicating these
>self-narratives.
>
>- Lester [7] has built believable agents for educational
environments
>that take into account pedagogical and believability constraints
when
>sequencing their behaviors. Similar in spirit to the work of
Sengers,
>these agents concern themselves with how they will be interpreted
by
>human observers when selecting behaviors. Lester has collaborated
>extensively with designers, artists, and animators in building his
>agents. He recognizes that the representational skills of artists
are
>important in building agents whose behavior can be interpreted by
>humans.  Recently, Lester has also begun a project in interactive
>fiction.
>
>
>Narrative Intelligence
>
>Generalizing from this collection of work allows us to begin
defining the
>concerns of Narrative Intelligence (NI). At its most fundamental,
NI
>is concerned with narrative as a way of understanding the
>world. Turning experience into a story is considered by many NIers
a
>fundamental mode of sense-making. This idea has a long tradition in
>narrative psychology as well as in AI (see, for example, the work
of
>Roger Schank [8]).
>
>In various strands of NI work, several themes emerge:
>(1) *Narrative as a system design principle*: If humans often make
sense
>of the world by assimilating it to narrative, then it makes sense
to
>design our systems so as to allow people to use their well-honed
>narrative skills in interpreting the system. In building systems
whose
>primary purpose is to tell a story, such as Mateas' or Elliott's
>work, this seems obvious. But the work of Sengers and Lester points
to
>the utility of using narrative to structure the visible behavior of
>autonomous agents, even if the primary function of the agent is not
to
>tell a story. Narrative design in this sense becomes a general
>principle when building any interactive system. Such ideas have
also
>been discussed in the human-computer interaction (HCI) literature,
>such as in the work of Don [4] or Laurel [6].
>
>(2) *Narrative as a way of structuring agents*: Conversely, since
>humans make sense of the world via narrative, perhaps artificial
>systems should also employ narrative in understanding the world.
Thus,
>for example, two artificial agents may want to organize their
>experience as stories they can share. Dautenhahn's and Nehaniv's
work
>points in this direction.
>
>(3) *Narrative as cross-disciplinary*:  If narrative is indeed, as
>many argue, a fundamental organizing principle of human experience,
>then it is unsurprising that many different fields to have an
interest
>in narrative. Work in NI has drawn on conceptions of narrative from
>many of these sources: *art*, in which narrative is understood as a
>form of representation; *psychology* (especially narrative
>psychology), in which narrative is thought of as a way in which
huamns
>make sense of the world; *cultural studies*, in which narrative is
>studied as a way in which a culture structures and propagates
>knowledge.
>
>Scope and questions of the symposium
>
>One of the major goals of this symposium is to bring together
>participants from different areas both within and without AI who
share
>a common interest in the intersection of narrative and AI. Within
AI,
>this symposium would solicit work from, but not limited to, the
>following areas:
>* Story understanding
>* Story generation
>* Narrative structure in interface design
>* Narrative structure in the design of autonomous agents
>* Believable agents (insofar as they participate in narrative
structure)
>* Interactive story-telling (includes interactive drama and
>interactive fiction)
>
>In addition, because NI researchers have drawn deep inspiration
from
>concepts of narrative from other disciplines, we hope to broaden
and
>solidify our understanding of narrative by including several
>participants from other research traditions, including:
>* Narrative psychology
>* Narrative theory
>* Art
>* Cultural studies
>Submissions from outside AI will be selected according to their
>relevance for illuminating the connections between narrative and
AI.
>
>This symposium hopes to address questions such as the following:
>* What role does or should narrative play in the design of
autonomous
>agents?
>* What models of narrative are appropriate in the design of
>interactive systems?
>* Is the appropriateness of a model of narrative a function of the
purpose
>of the interactive system? If so, what associations exist between
models of
>narrative and interactive systems?
>* How do humans make narrative sense of the world and of
interactive
>systems?
>* What sorts of narrative experiences are possible in interactive
>story-telling?
>* What AI architectures are appropriate for generating and
understanding
>narrative structure?
>* What role does or could narrative play within AI research?
>
>Structure of the symposium
>
>Submissions of a length of 3-5 pages will be solicited.  A subset
of
>these submissions will be selected for presentation to the group.
>One of the criterion of selection will be ensuring that a sample of
NI work
>from different disciplinary perspectives is presented.
>
>At the beginning of the symposium, a sheet will be handed out
>providing one paragraph descriptions of the work and interests of
all
>symposium goers (whether they are presenting or not). This will
>hopefully facilitate one-on-one conversation by allowing people to
>find others with similar interests.
>
>Time will be set aside for breakout groups to discuss specific
>topics. A set of potential topics will be created by the organizing
>committee in response to the submissions received. The actual
topics
>chosen will depend on the interests of the people at the symposium,
as
>indicated by their presentations and questions. The breakout groups
>will present a summary of their discussions to the whole
>group.
>
>Finally, a demo time will be set aside for people who have systems
>they would like to show.
>
>Bibliography
>[1] Bruner, J. The narrative construction of reality. Critical
Inquiry 18,
>1 (1991), 1-21.
>[2] Dautenhahn, K., and Nehaniv, C. Artificial life and natural
stories. In
>International Symposium on Artificial Life and Robotics (AROB III)
(Beppu,
>Oita, Japan, 1998), vol. 2, pp. 435-439.
>[3] Domike, S., Mateas, M., Vanouse, P. The recombinant history
apparatus
>presents: Terminal Time. Submitted to the Center for Twentieth
Century
>Studies, Milwaukee, WI, for inclusion in a book.
>[4] Don, A. Narrative and the interface. In The Art of
Human-Computer
>Interface Design, B. Laurel, Ed. Addison-Wesley, Reading, MA, 1990,
pp.
>383-391.
>[4A] Dyer M.G., Wolf T.C., Korsin M.: Boris - An In-Depth
Understander
>of Narratives, in Proceedings of the 7th International Joint
>Conference on Artificial Intelligence, University of British
Columbia,
>Vancouver, Canada, 1981.
>[5] Elliott, C., Brzezinski, J., Sheth, S., and Salvatoriello, R.
>Story-morphing in the affective reasoning paradigm: Generating
stories
>semi-automatically for use with 'emotionally intelligent'
multimedia
>agents. In Proceedings of the Second International Conference on
Autonomous
>Agents (New York, May 1998), K. P. Sycara and M Wooldridge, Eds.,
ACM
>Press.
>[6] Laurel, B . Computers as Theatre. Addison-Wesley, Reading, MA,
1991.
>[7] Lester, J., Stone, B. Increasing believability in animated
pedagogical
>agents. In Proceedings of the First International Conference on
Autonomous
>Agents, February 1997, W. Lewis Johnson, Eds., ACM Press.
>[8] Schank, R. Tell me a story: A new look at real and artificial
memory.
>Scribner, New York, 1990.
>[9] Sengers, P.  Narrative Intelligence.  In: Human Cognition and
>Social Agent Technology.  Ed. Kerstin Dautenhahn.  John Benjamins
>Publishing Company, 1999 (Forthcoming).
>[10] Sengers, P.  Anti-Boxology: Agent Design in Cultural Context.
>PhD Thesis, Carnegie Mellon University. CMU-CS-98-151.  August,
1998.
>
>
>Organizing Committee:
>
>Phoebe Sengers, co-chair
>Center for Art and Media Technology (ZKM)
>Institute for Visual Media
>Postfach 6909
>D-76049 Karlsruhe
>Germany
>phone +49 (0)721-8100-1500
>email phoebe@zkm.de
>
>Michael Mateas, co-chair
>Department of Computer Science
>Carnegie Mellon University
>5000 Forbes Ave.
>Pittsburgh, PA 15213
>USA
>phone (412) 268-3070
>email michaelm@cs.cmu.edu
>
>Kerstin Dautenhahn
>Department of Cybernetics
>University of Reading
>Whiteknights, PO Box 225
>Reading
>RG6 6AY
>UK
>phone +44 (0) 118-931 8218
>fax +44 (0) 118-931 8220
>email kd@cyber.reading.ac.uk
>
>Clark Elliott
>Institute for Applied Artificial Intelligence
>School of Computer Science, Telecommunications, and Information
Systems
>DePaul University
>243 South Wabash Ave.
>Chicago, IL 60604
>USA
>email elliott@elliott.cs.depaul.edu
>
>James Lester
>Department of Computer Science
>North Carolina State University
>Raleigh NC 27695-7534
>USA
>phone (919) 515-7534
>fax (919) 515-7925
>email lester@csc.ncsu.edu
>
>Chrystopher Nehaniv
>Interactive Systems Engineering
>Faculty of Engineering & Information Sciences
>University of Hertfordshire
>College Lane
>Hatfield, Hertfordshire AL10 9AB
>UK
>phone +44-1707-284-470
>fax +44-1707-284-303
>email c.l.nehaniv@herts.ac.uk
>
>
>Kathleen Biddick
>Stanford Humanities Center, Mariposa House, 546 Salvatierra Walk,
Stanford
>U., Stanford CA 94305-8630
>TEL: 650-725-1516; FAX: 650-723-1895; HOME: 650-321-5061
>