[Apologies for cross-postings]
Call for papers:
Probability Theory or Not?
Practical and Theoretical Concerns on Uncertainty Handling in AGI
AGI'13 Workshop: Uncertainty in Artificial General Intelligence
Location: Beijing, China
Date: July 31, 2013 (right before IJCAI)
http://www.agroparistech.fr/mia/agi-workshop/
Submission deadline: May 15, 2013
== Invited Speakers ==
*) Marcus Hutter
Research School of Computer Science
Australian National University, Australia
*) TBA
== Background ==
In the emerging field of AGI, more and more researchers have realized the
necessity of processing the uncertainty in the environment and the systems
in a systematic and well-justified way. Many people choose to use the most
developed mathematical theory, probability theory, to represent and process
the uncertainty involved, although alternatives exist. Various mathematical
models have been proposed, and many engineering designs also include
probabilistic or statistical calculations in them.
Though probabilistic models are popular, there are nevertheless some
fundamental issues to be resolved in these models, including
* What is measured by the “probability” in a model? Is it an objective
property of an event, a subjective strength of a belief, or a logic
relation between evidence and conclusion?
* How to evaluate the existing approaches to uncertainty handling in
AGI? Are the theoretical models applicable? Are the engineering models
justifiable?
* What are the basic requirements for uncertainty handling in AGI
systems to be legally applied? How can these requirements be satisfied
within real-world computational constraints? How well are these
requirements satisfied in AGI models?
* What is the most promising way to process uncertainty in AGI? To use
the classical model of probability theory, an extended or modified
probability theory, or another theory entirely?
By addressing the above issues during this AGI-13 workshop, we hope to
bring the theoreticians and the engineers to work together to develop
formal models that can be implemented in computer systems, so as to base
the AGI system on a solid foundation.
== Submission details ==
Papers can be submitted as either 2 pages extended abstracts or 7 pages
papers, in the AGI format. The submission deadline is May 15, 2013. All
papers should be written in English, and submitted as a PDF file to matt
dot ikle at gmail dot com.
Papers will be published on the website, but will not be included in the
main conference proceedings. An extended version of some selected papers
may be considered for a special issue in JAGI after the workshop. (Note
that JAGI has its own review process.)
== Workshop Organizers ==
Matthew Iklé, Adams State University, USA
Laurent Orseau, AgroParisTech, France
--
Laurent Orseau
Associate Professor in Artificial Intelligence
AgroParisTech, 16 rue Claude Bernard, 75005 Paris, France
Tel: +33 1 44 08 16 79
http://www.agroparistech.fr/mia/orseau
_______________________________________________
uai mailing list
uai@ENGR.ORST.EDU
https://secure.engr.oregonstate.edu/mailman/listinfo/uai
