perm filename MORE[F82,JMC]2 blob
sn#688555 filedate 1982-12-06 generic text, type C, neo UTF8
COMMENT ⊗ VALID 00004 PAGES
C REC PAGE DESCRIPTION
C00001 00001
C00002 00002 more[f82,jmc] More on Circumscription
C00003 00003 Each of the several papers that introduces a mode of non-monotonic reasoning
C00010 00004 w
C00017 ENDMK
C⊗;
more[f82,jmc] More on Circumscription
Abstract: (McCarthy 1980) introduced circumscription as a method
of non-monotonic reasoning. The present paper contains a new
version of circumscription based on the minimization of formulas
rather than predicates, the notion of prioritized circumscription,
some progress towards heuristics for the use of circumscription
and a discussion of several questions that have arisen concerning
circumscription and other modes of non-monotonic reasoning.
The present paper is logically self-contained, but the motivational
remarks of the previous paper are not repeated or superseded.
Each of the several papers that introduces a mode of non-monotonic reasoning
seems to have a particular application in mind. Perhaps we are looking
at different parts of an elephant. Here is a typology of non-monotonic
reasoning about which I solicit comments. This is intended as part of
a larger paper. The orientation is towards circumscription, but I suppose
the considerations apply to other formalisms as well.
Circumscription may have several uses.
1. As a communication convention. Suppose A tells B about
a situation involving a bird. If the bird may not be able to fly, and this
is relevant to solving the problem, then A should mention
the relevant information. Whereas if the bird can fly, there is
no requirement to mention the fact.
The circumscriptions to be made by the recipient of the communication
are those described in this paper. I have not yet considered how
the sender computes what to say and how this refrains from saying
that the bird can fly - presumably without any explicit decision.
2. As a database or information storage convention. It may be a
convention of a particular database that certain predicates have
their minimal extension. This generalizes the closed world
assumption. One such would be nnly-can-fly in the case of information
about birds.
Neither 1 nor 2 requires that most birds can fly.
Should it happen that most birds that are subject to the communication
or about which information is requested from the data base cannot fly, the
convention may lead to inefficiency but not incorrectness.
3. As a rule of conjecture. This use was emphasized
in (McCarthy 1980). The circumscriptions may be regarded as expressions of some
probabilistic notions such as "most birds can fly" or they may be
expressions of simple cases. Thus it is simple to conjecture that
there are no relevant present
material objects other
than those whose presence can be inferred. It is also
a simple conjecture that a tool asserted to be present is usable
for its normal function. Such conjecture sometimes conflict, but there
is nothing wrong with having incompatible conjectures on hand. Besides
the possibility of deciding that one is correct and the other wrong, it
is possible to use one for generating possible exceptions to the other.
4. As a representation of a policy. The example is Doyle's "The meeting
will be on Wednesday unless another decision is explicitly made".
5. As a very streamlined expression of probabilistic information when
numerical probabilities, especially conditional probabilities, are
unobtainable. Since circumscription doesn't provide numerical probabilities,
its probabilistic interpetation involves
probabilities that are either infinitesimal, within an
infinitesimal of one, or intermediate - without any discrimination
among the intermediate values. The circumscriptions give conditional
probabilities. Thus we may treat the probability that a bird
can't fly as an infinitesimal. However, if the rare
event occurs that the bird is a penguin, then the conditional probability that
it can fly is infinitesimal, but we may hear of some rare condition
that would allow it to fly after all.
Why don't we use finite
probabilities combined by the usual laws? That would be fine
if we had the numbers, but circumscription is usable when we can't
get the numbers or find their use inconvenient. Note that the
general probability that a bird can fly may be irrelevant, because
we are interested in particular situations which weigh in favor or
against a particular bird flying. Of course, circumscription
does not provide a means of weighing evidence; it is appropriate
when the information permits snap decisions. However, many
cases nominally treated in terms of weighing information are in fact
cases in which the weights are such that circumscription and other
defaults work better.
6. We might also speculate that certain laws of common sense physics
or common sense psychology is inherently non-monotonic or, more specifically,
involves circumscription. The speculation is that this common sense
information has some inherently preferred form.
Six different uses for non-monotonic reasoning seems
too many, so let's see if we can condense them. Maybe
the first two and even the fourth are linguistic conventions.
w
Job Queue PPN TV Line Jobnam Size Time lng.PL.now Seg SWR Alias
2 IOWQ⊗ 1MMD 23 TELNET 24P 0:02
3 STOP 1TOB 127 67 MAIL 56P 0:45
4 RUNQ 1SID 72 PACWAR 90P 13:34 17% 17% 11%
5 IOWQ 1HHB 73 70 MAIL 50P 0:03 1% 1%
10 IOWQ 1BCM 101 64 E 16P 0:02 1
13 IOWQ 1LES 24 E 24P 0:10 1
16 IOWQ 1JDH 150 75 E 24P 0:05 1 204JDH
19 STOP 1LER 116 110 DO 24P 1:40 CANSYS
21 IOWQ 1 YM 1 E 20P 0:02 2% 1
22 IOWQ 1LDG 123 63 TABLET 8P 0:05 11% 9% 8
28 STOP 1 ME 120 65 E 24P 2:16 1
29 IOWQ 1CAB 134 76 E 20P 0:33 1
31 STOP 1 YM Det E 18P 0:04 1
32 INTW 1JMC 140 73 E 16P 0:15 2% 1% 1 F82JMC
9 RUNQ- 1JMC 162 WHO 14P 0:00 F82JMC
36 STOP 1FWH 35 E 30P 0:35 1 PAPFWH
39 STOP 1JJC 75 77 FINGER 30P 1:39
45 STOP 1HDG 154 62 COPY 14P 0:17 1%
46 IOWQ 1LGC 15 E 28P 0:44 1 ATLGC
47 IOWQ 1 OK 106 101 E 16P 0:12 1
48 RUNQ 1RJB 104 PACWAR 90P 35:56 19% 21% 21%
51 IOWQ S82RWW 76 66 SPIDER 84P 0:09
53 IOWQ 1SCS 111 103 PACWAR 90P 31:58
6 IOWQ ACTSYS Det *SPY* 12P 20:41
7 INTW SLSYS Det *-SL-* 30P 0:06
11 IOWQ PNYACT Det PONY 58P 3:34
12 INTW SPLSYS Det [XSPL] 26P 10:16 1% 5%
14 IOWQ TEXSYS Det [TEX!] 242P 3:31 18
15 INTW RMDSYS Det WHOPHN 28P1:26:29 1%
17 INTW ETHSYS Det PUPSRV 24P 7:28
23 IOWQ 100100 Det COPY 14P 0:14
24 IOWQ* 100100 Det SPOOL 16P 0:06
25 INTW SPLSYS Det DOVER! 58P 13:16
33 INTW⊗ NS NS Det [-NS-] 4P 0:38
52 INTW⊗ ETHSYS Det ARPSER 20P 0:03
Job Segnam Size # Jobs
1 E 60P 11
8 W TABLET 16P 1
18 W HITEX 164P 1
Up Time: 11 days, 6 hours, 11 minutes
Null Time: 54% 0%
Wasted: 0% 0%
Core: 1566P 16P*
Usable: 3809P 3809P
Running Jobs: 1, 97P
Disk Queue: 0
DSK 21 BBOARD TXT 2 2 510 397 R
DSK 22 000554 026 1LDG 0 1 W
DSK 33 !L"/[R TEM NYT NS 0 1 RA
DSK 32 MORE F82JMC 10 11 RE
DSK 4 WAR COM 1SID 8 3 R
DSK 4 EVOL PCW 1SID 0 1 W
DSK 19 CPRINT DO CANSYS 1 2 RE
DSK 47 ARTICL TEX 1 OK 18 12 RA
DSK 29 SCIAM3 1CAB 128 55 R
DSK 33 !L"/V2 TEM AP NS 3 4 RA
DSK 2
DSK 2
DSK 10 MAIL$E TMP 1BCM 3 4 RE
DSK 48 WAR COM 1RJB 8 9 RE
DSK 48 EVOL PCW 1RJB 3 4 W
DSK 14 <TEX!> TXT TEXSYS 21 577
DSK 13 MSG MSG 1LES 150 151 RAE
DSK 16 SOL3 TEX 204JDH 3 4 RA
DSK 46 CSREAS ATLGC 143 37 RA
DSK 6 CRASH SAV ACTSYS 19 20 RA
DSK 14 TEX PSC TEXDEK 459 3 R
DSK 11 DEC 82 PNYACT 6 7
DSK 24
DSK # 0 [SWAP] SYS 1 4 0 24193
UDP2 # 3
PUP 52
PUP 17
IMP 2
IMP 52
IMP 33
TTY162 9
TTY14 # 0
TTY16 33
TTY17 33
TTY30 11
↑C
.