COMMENT ⊗   VALID 00003 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	.require "memo.pub[let,jmc]" source
C00004 00003	.if false then begin
C00005 ENDMK
C⊗;
.require "memo.pub[let,jmc]" source
.cb TWO LECTURES ON FORMALIZED NON-MONOTONIC REASONING

	
	Logical deduction is monotonic in that if a sentence ⊗p 
is deducible from a collection ⊗A of sentences which is contained in
a larger collection ⊗B, then ⊗p is deducible from ⊗B.  Human reasoning
is often non-monotonic; if you know that I have a car you may conclude
that I will drive home, but on learning that the car is broken you
will draw a different conclusion.  Recently McCarthy, McDermott and
Doyle, and Reiter have developed different formalizations of non-monotonic
reasoning for artificial intelligence purposes.  The lecture will
discuss these formalisms together with many examples of non-monotonic
reasoning.

Time: Wednesday, May l4, 2:30 to 4:00 p.m.                     
      Time to be determined for second lecture.

Place: Room 402, Margaret Jacks Hall

Speaker: John McCarthy
.if false then begin
ANNOUN[S80,JMC] is a draft announcement of two lectures I plan to give.
However, please check on room availability and whether it conflicts with
seminars in computer science or philosophy.  I guess a bigger room than
301 MJH is needed.

Notices should go to philosophy department, Professors Barwise and Kreisel
personally, CS bulletin boards.  Also via computer mail to Bobrow%MAXC and
Nilsson%SRI with a note suggesting that they post it if they think people
will be interested.  Also by computer mail to AIPHIL[F79,JMC].
.end