perception of rapid progress

CSGnet Archive CSG1992
1

[From: Bruce Nevin (Tue 920804 11:26:03)]

It seems a truism that people are not interested in alternatives (viz.
PCT) so long as they perceive themselves as "getting somewhere" by their
present means. Here's a perspective on a bunch of people who believe
they're progressing toward their goals in robotics and artificial life.
How could the news that they're not be framed in such a way that they
see it as a contribution (mid-course correction) rather than a demand
that they give up and do something else (abandon ship)?

If you receive cybsys-L mail, my apology for repeating the following:

···

From: Cliff Joslyn <cybsys@bingsuns.cc.binghamton.edu>
Subject: ALife III Conference Report, Hugo de Garis, ETL, Japan.
To: Multiple recipients of list CYBSYS-L
<CYBSYS-L%BINGVMB.BITNET@pucc.Princeton.EDU>

Really-Really-From: degaris@etl.go.jp (Hugo de Garis)
Really-From: alife@cognet.ucla.edu (Artificial Life Digest)

[ The following is a cross-post from Artificial Life Research List
Digest Number 081 Monday, August 3rd 1992. You are encouraged to
subscribe by sending mail to alife-request@cognet.ucla.edu - Moderator ]

Dear ALifers,
                Here is a quick report on the highlights of the ALife III
Conference which was held in Santa Fe, New Mexico in June 1992. It reflects the
personal biases and interests of the author.

Far and away the best paper was by Gerald Joyce, who spoke on his "evolution of
molecules" work. He takes RNA, and can mutate one, two or more specific bases,
and then clone the mutants in huge numbers. These mutants are then subject to
a selectionist test, so that only the more successful mutants survive.
The survivors are then further mutated etc, until molecules are evolved which
perform some desired function. This is probable future Nobel work. Good luck
Gerald.

John Koza used his Genetic Programming technique (evolution of Lisp programs)
to evolve self reproducing systems, and told his audience that the size of the
search space (with his primitives) was only of the order of a billion or so.
This is exciting, because its a lot less than earlier estimates (eg von Neuman's
29 state reproductive cellular automata Turing machine). It means that
it will probably be possible in a year or so to evolve selfreproductive
systems which can also do something useful. This will be essential when nano
tech is finally with us. Nanoscale machines will have to self reproduce in
order to build macroscale systems.

The biological robot (biot) community went home with the message that an
evolutionary approach to building biot nervous systems is the way to go.
Brooks-style handcrafting has become too complex and needs an evolutionary
approach. Even Brooks was talking about evolving gnat robots.

Dave Ackley showed that Lamarkian evolution can "blow the doors off" standard
GA evolutionary learning. An amusing and engaging talk.

L. Buss and W. Fontana teamed up to present a mathematical theory of the
development of systems towards life like behavior. They presented levels
0, 1, and 2, with increasing sophistication, e.g. self replication, then
genomes, etc. It was hard to follow but felt important. At the end of the
talk they claimed they had implemented it all in a computer program but
they failed to present results (and this was to an audience of
70% computer types!).

A fast-talking half-crazy Canadian (M. Tilden) upstaged Rod Brooks by showing
that he could do a lot of what Brooks does, but for a thousandth of the
price. Tilden makes ultra cheap little robots with an amazing functionality.
I hope Rod Brooks went home with the lesson. Funders take note!
Tilden will go down in ALife history for his reply to a non native english
speaker's question, "Why do you talk so fast?". Tilden's
reply was, "Whydoyoulistensoslowly?!"

J. Smits made tiny bilayer silicon strips curl up when a current is applied.
He intends to use these strips as mechanical movers for "silicon ants".

Randy Beer now uses GAs to build his insect circuits. Darwinian Robots was
one of the themes of the conference.

Maja Mataric (Brooks student) presented a video on Brooks version of swarm
intelligence. 20 robots were supposed to perform behaviors by emergence.
The video angles were too low to see effectively what was happening. Maybe
she should have done the filming in a basketball court and taken the shots
from a ladder.

Personally, I went away all fired up that "evolvable hardware" is possible.
I learned about FPGAs (software configurable hardware), so it will be possible
to treat the configuration bitstrings of FGPAs as chromosomes in a GA. Thus the
technology may exist today to fulfill my dream of building "Darwin Machines".

I just got back to Japan, so this report probably reflects my jet lag.

Cheers,

         Hugo de Garis,
         Electro Technical Lab (ETL), Japan.
-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-=+=-

Bruce
bn@bbn.com

2

Re : Bruce Nevin (920804)

It seems a truism that people are not interested in alternatives (viz.
PCT) so long as they perceive themselves as "getting somewhere" by their
present means. Here's a perspective on a bunch of people who believe
they're progressing toward their goals in robotics and artificial life.
How could the news that they're not be framed in such a way that they
see it as a contribution (mid-course correction) rather than a demand
that they give up and do something else (abandon ship)?

my immediate reaction to this, i'm afraid, was to rant a little.
i will not, both for the sake of propriety, and to defend myself
at the outset from what unfortunately i imagine would be the response...
something along the lines of 'see... resistance to new ideas is so strong
that a suggestion of change brings forth only anger, not re-examination.'

first... what does mr nevin suppose the goals of artifical life research
to be, exactly?

second... i think that there is a significant difference in the usage and
examination of 'behavior' from the points of PCT and artificial life.
i'm not sure how to express it, but it occurs to me as a difference between
thought and action (sorry for the trite terminology, but i'm not feeling
particularly articulate right now). i found mr powers' statement in BCP
that he was after, not a model of behavior, but a model of how an organism
might be organized that it may behave, as a fundamental. both an insight
and a demarcation as to what PCT's realm of competence is. kuhn, you may
remember, says that one of the functions of a science's self-declaration
as a science is to *limit* its universe of discourse... to provide guidelines
for what it is *not* about. as an interested observer and dilletante, i see
no particular conflict between CT and Alife that should require one to
impose on or seek out the other at this time.

third... occam is frequently invoked here. CT claims to offer better
explanatory power than the current mainstream of psychology, with less
recourse to mumbo. 'does it work' is the mantra, as it should be in a
field that derives, as CT does, from an engineering perspective. well then,
the razor cuts both ways. Alife research is developing models that function.
tools are being made which accomplish computational tasks. i really
don't see any reason why its practitioners should modify their approach,
since they are in fact (not just appearance) progressing toward their goals.

-----------< Cognitive Dissonance is a 20th Century Art Form >-----------
Eric Harnden (Ronin)
<HARNDEN@AUVM.BITNET> or <HARNDEN@AMERICAN.EDU>
The American University Physics Dept.
4400 Mass. Ave. NW, Washington, DC, 20016-8058
(202) 885-2748
---------------------< Join the Cognitive Dissidents >-------------------

···

From: Eric Harnden (920805)

3

(From Rick Marken (920805)

I would like to add my voice to those of Bill Powers (920805.1000)
and Bruce Nevin (Wed 92085 12:21:44) who asked Eric Harnden (920805)
to explain what ALife is about. Bruce wrote a particularly articulate
reply to Eric, who seemed to take umbrage at Bruce's implication that,
given its current assumptions, ALife was not going reach the goal of
producing models that act like living systems.

Bruce said (to Eric):

I believe ALife efforts will lead to products that are in fundamental
ways unlike living things because it appears to me extremely likely that
living things are perceptual control systems, as PCT claims, yet I see
no indication that ALife entities control input perceptions relative to
internally-held reference perceptions by way of effectors through the
entity's environment, not even as emergent properties of those entities.
Hence, the claim that they are not really achieving the goals that they
think they are (entities that are like living things, or even that act
like living things consistently amid disturbances).

If you can show that I am wrong in the second part of this (the claim
about ALife), then rapprochement should be easy. If you cannot show
that this is wrong, then I see two choices for you (there may be more).
You might seek to reconcile the conflict by deciding that you must
persuade us that the first part of this is wrong (the claim about living
things being control systems). Or you might decide that persuading a
bunch of fanatics, who are on the fringes of where the real action is
anyway, is not worth the trouble, and just go off and forget about PCT.

I agree with Bruce's analysis and would very much like to see a reply to
it by someone who is really involved in ALife, as Eric seems to be.

Bruce again:

I see only one way that you can conclude from all of this that there is
"no particular conflict between CT and Alife that should require one to
impose on or seek out the other at this time," and that is that you must
understand ALife as modelling behavioral outputs; or at least you must
believe that ALife does not aim to create models whose behavior (as
byproducts of control) is like that of the organisms modelled. But this
is just the claim about ALife that I invited you to demonstrate was
wrong.

Beautifully said!

If Eric would answer Bruce's (and Bill's) questions about ALife it would,
indeed, help all of us who are trying to understand how living systems
work (I assume that that includes most of us on csg-l).

Thanks

Rick

···

**************************************************************

Richard S. Marken USMail: 10459 Holman Ave
The Aerospace Corporation Los Angeles, CA 90024
E-mail: marken@aero.org
(310) 336-6214 (day)
(310) 474-0313 (evening)