🚀 go-pugleaf

ACM SIGART Chicago
December, 1995
News of the Special Interest Group on Artificial Intelligence of the Chicago
Chapter of the Association for Computing Machinery

NEWS AND NOTES

Please note upcoming meeting dates on your calendar now so you wont miss a
single meeting:
        December 7, 1995 - Richard Neapoltan, Can a Computer Learn a Belief
                           System.
        March 3, 1996 - Joint meeting with INFORMS. Panel discussion scheduled.
        June 6, 1996 - Dale Reed

This months talk presents some compelling ides on the subject of creating
thinking machines.  A Belief System could become the central core of the next
generation of expert systems.

Special thanks to Frank Rinaldo for his presentation on AI and the Financial
Markets at the September meeting.

DECEMBER MEETING ANNOUNCEMENT:
-----------------------------------------------------------

Can a Computer Learn a Belief System?
   by Richard Neapoltan

Abstract:

Can we develop a system that learns a belief system in a lifelike way?

Since 1970 the Artificial Intelligence community has focused on developing
domain-specific knowledge-based systems, which are systems that have the
expertise of humans in specific domains. We have achieved good success in this
area.  However, if we require that an intelligent agent be capable of rapidly
interpreting information in a useful time frame and be able to modify its
behavior based on changes in the environment, we have made little progress in
the development of artificial intelligence.  The current trend in the AI
community is towards further development of normative systems.  The development
of an entity with artificial intelligence seems to have been abandoned.

While the AI community has focused on knowledge-based systems, some biologists
and philosophers have taken a different approach to AI, which could be called
the artificial life approach.  These researchers are searching for algorithms
that describe the process by which life self-constructs by interacting with
environment.  So far the artificial life approach has been applied to developing
entities that learn crude skills such as seeking light and not bumping into
walls.

The belief network is one of the most important structures for representing
knowledge in the knowledge-based systems mentioned above. Some AI researchers
maintain that humans structure causal knowledge in a belief (causal) network.
This is in agreement with some philosophers and psychologists who maintain that
the cause-effect relation is the fundamental unit of human knowledge.  However,
these AI researchers have not addressed how the human learns his belief network.
The algorithms they've developed for learning belief networks are based on
normative principles and do not learn incrementally.  Again, the focus has been
to develop knowledge-based systems rather than artificial intelligence.

This talk describes the belief network, summarize the arguments that human
knowledge is structured in a belief network, describes some of the artificial
life algorithms, and addresses the question as to whether we can use these or
other algorithms to learn a belief network.  That is, can we combine the methods
of the Artificial Intelligence and Artificial Life communities to develop
systems that learn high-level cognitive skills in a lifelike way? 


SPEAKER:

Richard Neapoltan is Professor of Computer Science at Northeastern Illinois
University. He has written over 20 papers on Uncertain Inference in Artificial
Intelligence, one of his major research interests.  Professor Neapoltan is the
author of Probabilistic Reasoning and Expert Systems, published in 1990, by
Wiley.  His new book Foundations of Algorithms, published by Heath, is coming
out in January of 1996.  Look for it!


Meeting Details

Date: Thursday, December 7, 1995
Time: 5:30 p.m. Refreshments & Snacks; 6:30 p.m. Speaker
Place: IITs Chicago Kent Law School  (Room 355)
          565 West Adams St., Chicago
Cost: $8.00 Members; $10.00 Non-Members; $4.00 Students
Reservations: Call John Garton at Byte Knights, (312) 327-9250. 
Please DO try to  make a reservation, it really helps us plan 
the meeting space.
Information: Call John Garton, (312) 327-9250.

Travel Directions to SIGART Meeting
Via Metra:      Use Union Street Station.  Walk West on Adams St.
Via CTA:        Take Ravenswood, Evanston, or Midway trains to Adams St.  Walk
                  West on Adams St.
Via Bus:        41, 60, 125, 151, 156, 157 to Union Station.  Walk West on Adams
                  St.
Via Car:        Kennedy/Dan Ryan to Monroe exit.  Go East to Clinton, South to
                  Adams.  On street parking is limited.  
                  There are several parking lots in the area.
Walking:        Walk West from the Loop on Adams.

Chicago Chapter ACM
PO Box 2381
Chicago, IL  60690

--