AI, The Fundamental Social Aggregation Challenge, and the Autonomy of Hybrid Agent Groups

  • About Us
  • Gifts
  • AITopics
  • AI Magazine
  • Conferences
  • Library
  • Membership
  • Publications
  • Symposia
  • Contact

AI, The Fundamental Social Aggregation Challenge, and the Autonomy of Hybrid Agent Groups

Papers from the 2012 AAAI Spring Symposium

W. F. Lawless, Don Sofge, Mark Klein, Laurent Chaudron, Program Cochairs

Technical Report SS-12-01
74 pp., $30.00
ISBN 978-1-57735-550-2
[Add to Cart] [View Cart]

This AAAI Symposium has in mind an approach to the solution of the following, nonexhaustive list of questions for hybrid teams: (1) What makes a group autonomous? (2) Is individual autonomy possible for a single hybrid agent (Sharkey, 2008)? If so, can an autonomous individual agent be a social member of a team? What else might be different? (3) Related to question 2, why are hybrid agent reports of behavior not the same as observed behavior? (4) Why are groups, unlike individuals (Wickens, 1992), able to multitask effectively (Ambrose, 2001) and, from an information theory perspective, more efficiently (Conant & Ashby, 1970)? (5) Human systems perform differently depending upon whether they are organized and controlled centrally or locally (see Ahiedh, 2009; Hayek, 1994). This question addresses the theoretical perspective of information flow; for example, is information about a hybrid organization's defenses transmitted with natural language better than the information obtained from observing the same organization's performance under a social perturbation (such as organizational volatility as a consequence of a cyber-attack)? (6) As an alternative to question 4, is the rational construction of reality derived competitively at the individual level different from the dynamic information derived cooperatively at the organizational level? (7) Why isn't the production of hybrid social autonomy a simple problem? (8) What would a mathematical model of aggregation look like (Lawless et al., 2010)? Would such a mathematical model account for social autonomy; the inability of an agent's reports to capture its behavior; and the differences observed between the results obtained with the applications of game theory to the toy problems favored by researchers (viz., normative solutions) versus the real solutions found in the field made by autonomous organizations of humans (such as Apple, Google)?

AAAI Press

Ordering Information

Conference Reports

EAAI Symposium Reports

Fall Symposium Reports

Spring Symposium Reports

Workshop Reports

Press Books

Press Proceedings

Journals

For Authors

Permissions Requests

Other Links

AAAI Home Page

Awards

Calendar

Jobs

Meetings

AAAI Press

Resources

AAAI Workshops

Follow @RealAAAI

This site is protected by copyright and trademark laws under US and International law. All rights reserved. Copyright © 1995–2020 Association for the Advancement of Artificial Intelligence.
Your use of this site is subject to our Terms and Conditions and Privacy Policy | Home | About AAAI | Search | Contact AAAI
AAAI Conferences | AI Magazine | AITopics | Awards | Calendar | Digital Library | Jobs | Meetings | Membership | Press | Press Room | Publications | Resources | Symposia | Workshops