Machine Self-Reference and Consciousness John Case Depa rtment - - PDF document
▶
SMART_READER_LITE
LIVE PREVIEW
Machine Self-Reference and Consciousness John Case Depa rtment - - PDF document
Sep 07, 2023 347 likes •493 views
Machine Self-Reference and Consciousness John Case Depa rtment of Computer and Info rmation Sciences 103 Smith Hall Universit y of Dela w a re New a rk, DE 19716 USA Email: case@cis.udel.edu http://www.cis.udel.edu/
SLIDE 1Machine Self-Reference and Consciousness John Case Depa rtment
f
Computer and Info rmation Sciences 103 Smith Hall Universit y
f
Dela w a re New a rk, DE 19716 USA Email: case@cis.udel.edu http://www.cis.udel.edu/ case 1
SLIDE 2First Problem Can machines tak e their entire inter- nal mechanism into account as data? Can they have \complete self- kno wledge" and use it in their decisions and computations? W e need to mak e sure there is not some inherent pa rado x in this. 2
SLIDE 3
OF THEMSELVES? 1. ________
M
MODEL OF M MODEL OF MODEL OF M
. . .
INF. INFINITE REGRESS! HENCE, M NOT A MACHINE. THEREFORE, M CANNOT CONTAIN A MODEL OF ITSELF! _________ M CAN MACHINES CONTAIN A COMPLETE MODEL
3
SLIDE 4So | 2. Can machines create a mo del
f
themselves | exter- nal to themselves? YES! | b y: a. Self-Replication
r
b. Mirro rs. W e're gonna do it with mirro rs! | No smok e, just mirro rs. 4
SLIDE 5
3 + 4 = ? 172 123 x
The rob
t
has a transpa rent front so its internal mechanism is visible. It faces a mirro r and a writing b
a
rd, the latter fo r \calculations." It is sho wn having copied already a p
rtion
f
its internal mechanism, co r- rected fo r mirro r reversal,
nto
the b
a
rd. It will cop y the rest. Then it can do anything p reassigned and algo rithmic with its b
a
rd data consisting
f:
its complete (lo w-level) self-mo del and any
ther
data. The ab
ve
essentially depicts Kleene's Strong Recursion Theo rem (1936) from Computabilit y Theo ry (see [Cas94,RC94]). 5
SLIDE 6Levels
f
Self-Mo delling? The complete wiring diagram
f
a ma- chine p rovides a lo w-level self-mo del. Other, higher-level kinds
f
self- mo deling a re
f
interest, e.g., general descriptions
f
b ehavio ral p rop ensi- ties. A nice inhuman example (p rovided b y a machine) is: I compute a strictly increasing mathematical function. A human example is: I'm grump y , up
n
a rising, 85%
f
the time. F
r
machines, which w e lik ely a re [Jac90,Cas99
],
such higher-level self- kno wledge ma y b e p roved from some p
w
erful, co rrect mathematical theo ry p rovided the theo ry has access to the complete lo w-level self-mo del. Hence, the complete, lo w-level self- mo del is mo re basic.
The
exp ected b ehavio rs in a discrete, quantum mechanical w
rld
a re computable! 6
SLIDE 7Human Thought W e tak e the p
int
f
view that human thought inherently involves (attenuated) sensing in any
ne
f
the senso ry mo dalities. E.g., a. V
cal
tract \kinesthetic" [W at70] and/o r audito ry sensing fo r inner sp eech. b. There is imp
rtant
sha ring
f
b rain ma- chinery b et w een vision and p ro duction and manipulation
f
mental images. Many in- genious exp eriments sho w that the same unusual p erceptual eects
ccur
with b
th
real images and imagined
nes
[FS77,Fin80,She78,Kos83,CRS94]. In the follo wing w e will exploit fo r exp
sition
the visual mo dalit y since it admits
f
pic- to rially , metapho rically rep resenting the
ther
mo dalities. Generally the
nly
asp ects
f
ur
inner cog- nitive mechanism and structure w e humans can kno w b y conscious thought a re b y such means as: detecting
\silvering," the degree to which it lets light through vs. reects it. 8
SLIDE 9The Rob
t
Mo died A ttach, then, the tunable mirro r/b
a
rd to the transpa rent and senso ry front
f
the rob
t
to
btain
the new rob
t:
NewRobot Tunable Mirror/Board External Images
Int. Images
The new rob
t
controls ho w much it lo
ks
at externally generated data and ho w much it lo
ks
at internally gener- ated data, e.g, images
f
its
wn
mech- anism. The attached, tunable mirro r/b
a
rd is no w pa rt
f
the new rob
t.
9
SLIDE 10The Human Case? The rob
t's
tunable mirro r/b
a
rd is analogous to the human senso ry \sur- face." The latter is also tunable as to ho w much it attends to internal \im- ages" and ho w much it attends to ex- ternal. Ho w ever, w e humans can
nly
\see" the pa rt
f
ur
internal cognitive struc- ture
riginally
built from sense data and sent back to
ur
senso ry surface to b e re-exp erienced as mo died and, t yp- ically , attenuated, further sense data. This is not surp rising since w e lik ely evolved from sensing-only
rganisms.
Of course, any further depth
f
human cognitive structure
r
mechanism w e a re to lea rn ab
ut
will dep end
n
such w
nderful
a rticial techniques as scien- tic exp eriments, e.g., directly in neu- rophysiology , indirectly from the hu- man genome p roject, . . . . 10
SLIDE 11Lessons Of Machine Case? F rom Kleene's Recursion Theo rem (eventually) came
ur
mo died rob
t
with attached, tunable mirro r/b
a
rd. In applications
f
Kleene's Recursion Theo rem [Cas94,RC94] (within Com- putabilit y Theo ry) w e see that, while is it not needed to compute all that is computable, a. It p rovides very succinct p ro
fs
and p rogram constructs. b. F rom a game-theo retic viewp
int,
in some cases, a (machine) pla y er's self-kno wledge is an imp
rtant
comp
nent
f
its winning strategy . Quite p
ssibly
, then,
ur
wn,
less complete, human version
f
self- reection evolved thanks to a p remium
n
compact (i.e., succinct) b rains and the need to win survival games. 11
SLIDE 12Summa ry Kleene's Strong Recursion Theo rem p rovides fo r self-referential machines/p rograms. In eect, such a machine/p rogram externally p rojects
nto
a mirro r a complete, lo w level mo del
f
itself (i.e., wiring diagram,
w
cha rt, p rogram text, . . . ). W e analyzed this machine self-reference as an idealization
f
the self-mo deling comp
nent
f
human consciousness. Human self-mo deling is rather p
r
(but vastly b etter than self-mo deling in
ther
animals). Analyzed w as the manner in which the sepa ra- tion ab
ve
b et w een 1. the externally p rojected self-mo del and 2. the machine so-mo deled and doing this p rojection applies to the human case. The analog
f
the mirro r ab
ve
is the human senso ry \surface," tunable as to its degree
f
\silvering!" F rom applications
f
Kleene's Theo rem in Computabilit y Theo ry: complete machine self- mo deling aids with machine/p rogram suc- cinctness and with winning games. P erhaps the uses
f
consciousness a re simila r: need to have a compact b rain and to win survival games. 12
SLIDE 13References [Cas94] J. Case. Innita ry self-reference in lea rning the-
ry
. Journal
f
Exp erimental and Theo retical Arti- cial Intelligence, 6:3{16, 1994. [Cas99] J. Case. The p
w
er
f
vacillation in language lea rning. SIAM Journal
n
Computing, 1999. T
app
ea r. [CRS94] J. Case, D. Rajan, and A. Shende. Rep resent- ing the spatial/kinematic domain and lattice com- puters. Journal
f
Exp erimental and Theo retical Articial Intelligence, 6:17{40, 1994. [Fin80] R. A. Fink e. Levels
f
equivalence in imagery and p erception. Psychological Review, 87:113{ 139, 1980. [FS77] R. A. Fink e and M. J. Schmidt. Orientation- sp ecic colo r after-eects follo wing imagination. Journal
f
Exp erimental Psychology: Human P er- ception and P erfo rmance, 3:599{606, 1977. [Jac90] R. Jack endo. Consciousness and the Compu- tational Mind. Bradfo rd Bo
ks,
1990. [Kos83] S. Kosslyn. Ghosts in the Mind's Machine: Creating and Using Images in the Brain. Ha rva rd Univ. Press, Camb ridge, Massachusetts, 1983. [RC94] J. Ro y er and J. Case. Sub recursive Program- ming Systems: Complexit y and Succinctness. Re- sea rch monograph in Progress in Theo retical Com- puter Science. Birkh
auser
Boston, 1994. [She78] R. N. Shepa rd. The mental image. American Psychologist, 33:123{137, 1978. [W at70] J. W atson. Behavio rism. W.W. No rton, 1970. 13
