AAAI Spring Symposium Series 2005

Call for Participation

Persistent Assistants: Living and Working with AI


Consider a future in which intelligent agents play a significant role in our
personal and professional lives: smart houses will anticipate our actions and
needs, while personalized agents will tailor our entertainment to our
preferences, purchase goods for us on-line, monitor our health, remind us to
take medications, and even drive us to the store. At work, agents will
support our job roles via assistance that ranges from organizing meetings to
ensuring safety in complicated and stressful situations (such as in operating
nuclear power plants and conducting space missions). These examples have two
unifying features: they call on us to delegate decisions to an agent whose
behavior will materially affect our interests and/or our well-being, and they
require a close partnership between users and agents over an extended period
of time in order to get the job done.  

What will it take to enable such scenarios? What tasks can we best support?
As a class, effective assistants will need considerable human factors and
human interaction capabilities, as well as a capacity to acquire,
communicate, and employ potentially detailed cognitive models. We see many
opportunities for persistent assistants, which include: 

    Robotic exploration
    Robotic assistance
    Smart houses
    Elder care
    Personalized information retrieval
    Personal and enterprise information management
    SPAM filters
    Proxy agents
    Tutoring
    Games
    Software "wizards"
    Situation summarization 
    Cognitive prostheses
          
The desire to construct persistent assistants raises broad questions at the
intersection of the fields of autonomous systems and human centered
computing. For example: 

    + How does the context of persistent assistance shape user-agent
      interaction?  
    + What will make persistent interactions fundamentally constructive
      / collaborative? 
    + What requirements do particular tasks and user populations impose? 
    + How can a person communicate changing intentions, goals, and tasks to
      an agent? 
    + How can an agent communicate its understanding and intentions to a
      person?  
    + How is trust developed and maintained?  
    + What is the tradeoff between predictable behavior and adjustable
      autonomy? 
    + How can we implement systems that exhibit adjustable autonomy, acquire
      cognitive models, employ preferences to tailor behavior, and
      collaborate gracefully with people over time? 
    + How will persistent assistants affect people's social and interpersonal
      relations? 
    + How should mixed teams of many people and many agents interact? 

This symposium will bring together practitioners of artificial intelligence,
human computer interaction, cognitive modeling, robotics and assistive
technologies with a focus on helping people perform tasks more efficiently
and more safely. 

Topics of interest include:

    Plausible task domains 
    User modeling
    Adaptive interfaces
    Adjustable autonomy
    Learning over extended operation
    Human-agent collaboration 
    Technologies for establishing trust
    Intent recognition
    Voice recognition and interaction
    Task tracking
    Cognitive and behavioral modeling
    Planning, scheduling and constraint satisfaction
    
This symposium continues and extends the topics of the very successful Spring
Symposia on Human Interaction with Autonomous Systems in Complex Environments
(2003) and Human Interaction with Autonomous Systems over Extended Operations
(2004). It is intended to foster interactions among a highly
interdisciplinary set of participants by including presentations from
distinct perspectives, and by allocating ample time for discussions.  

Submissions

Papers should focus on the technical challenges of assisting people with
ongoing tasks, and/or on problem domains that require persistent assistants.
They can describe completed work, or work in progress. The organizing
committee will carefully review all papers. 

Submissions should be in AAAI format and no more than 8 pages in length.
Please send them by email(PDF preferred) to Daniel Shapiro (dgs@stanford.edu)
for distribution to the committee. The deadline for submissions is October 8,
2004.  

Organizing Committee:

Daniel Shapiro, ISLE/Stanford University (Chair), dgs@stanford.edu
Pauline Berry, SRI (Co-chair), berry@ai.sri.com
Jihn Gersh, Johns Hopkins University (Co-chair), john.gersh@jhuapl.edu
Nathan Schurr, University of Southern California (Co-chair), schurr@usc.edu
David Kortenkamp, NASA Johnson Space Center/Metrica Inc., korten@traclabs.com
Barney Pell, NASA Ames Research Center, barneypell@yahoo.com
Richard Simpson, University of Pittsburgh, ris20@pitt.edu

