<html>
<body>
At 09:46 AM 11/30/2004 -0800, Val Geist wrote:<br><br>
<blockquote type=cite class=cite cite>
<dl>
<dd>From:</b> <a href="mailto:HowlBloom@aol.com">HowlBloom@aol.com</a>
<br>
<dd>To:</b> <a href="mailto:paleopsych@paleopsych.org">paleopsych@paleopsych.org</a> <br>
<dd>Cc:</b> <a href="mailto:dranees@compuserve.com">dranees@compuserve.com</a> <br>
<dd>Sent:</b> Friday, November 26, 2004 8:18 PM<br>
<dd>Subject:</b> Re: [Paleopsych] genetics as an intelligent system<br><br>
<dd>Instead of "metagenetics", can I off<br>
<dd>When did the full suite of modern atoms--the 92 natural elements--become complete? Did the full panoply of modern atoms arrive after the collapse of the first meg-stars, stars that swelled, ignited, then died off very quickly? That first period of star death would have been a mere two million years after the big bang.<br>
<dd> </blockquote>
</dl><br>
On this one --- I am a heretic. Closer to the Vedas than to the<br>
Old Testament book of Le Maitre and Gamow.<br>
MAYBE the universe is not infinite in time... maybe not... but I<br>
do not believe in the recent Big Bang theories. Or superstrings for that matter.<br>
There is more empirical data for ... well, there is no empirical data at all for the latter.<br>
<br><br>
<br>
<blockquote type=cite class=cite cite>
<dl>
<dd> <br>
<dd>Paul's words suggest that restlessness and boredom have been a key part of this learning system. I've been calling this a restless cosmos, a driven cosmos, an obsessive compulsive cosmos for a very long time. But Paul is suggesting that we make computer-based learning machines restless too. That we make them try out new possibilities just for the hell of it, just to evade the pain of boredom, the pain of staying precisely the same, the pain of ennui. Paul is on the brink of suggesting that we make computational programs hunger for pop culture, for music and games that test and expand the silicon brain in new ways.<br>
<dd> <br>
<dd>Paul is suggesting that genes may be as restless and boredom-prone as Baudelaire, who painted ennui as the ultimate pain. He's suggesting that on the sly, when they're not working, genes play around and dance in leisure time. Or at least that's what Paul's ideas inspire in me.<br>
<dd> </blockquote>
</dl><br>
Basically right. Exploration is an unavoidable issue in designing true intelligent systems.<br><br>
<br>
<blockquote type=cite class=cite cite>
<dl>
<dd>I know that leisure, entertainment, pop culture, art, and play are not useless. I've known it since I began my 20 years of fieldwork in these fields--poetry, art, magazine publishing, and finally popular music. Paul seems to be whispering to me that these cultural expressions may be a stochastic search for new possibilities. And his words suggest to me that genes play games too. They play the sort of musical games--establishment of a theme, then variation on it--that Greg's mechanisms make possible.<br>
<dd> <br>
<dd>Paul, my apologies if I've bent your words, but they're extraordinarily evocative.<br>
<dd> <br>
<dd>Can you share the Einstein-Bear connection you were pondering?<br>
</blockquote>
</dl><br><br>
I have imagined a couple of pop stories or musical comedies I would want to write if<br>
I were in that business.<br><br>
One, on research... a musical comedy with a cheap paperback book version...<br>
called "Hot Topics." On the cover, a Dean and a new assistant dressed in red and<br>
a certain kind of unique deanish expression. (Not all deans have it but...)<br><br>
Inside... a big announcement to his faculty... that he has been looking for a new research thrust<br>
for the place... something truly new and exciting... and his assistant announces... "sex." One old<br>
guy grouses "right on one out of two...". <br><br>
And "hot topics" really do warp a lot of what goes on in universities. Better than being a living mummy I suppose,<br>
but there are other choices...<br><br>
==========<br><br>
Another image... probably a lot less marketable, but maybe more useful.<br><br>
A seance... the medium beings by warning that .. the astral world is full of spirits with <br>
unfinished business on earth but she will try to focus on the ones truly ... positive for<br>
the attendees... and then... a hush... and the voice comes out of her...<br><br>
"I am one of the spirits with enormous unfinished business on earth...<br>
my name is Albert Einstein... "<br><br>
One attendees says: "But you never even BELIEVED in ghosts..."<br><br>
"Yes, but that was before I became one. We all must bow to empirical data in the end.<br>
And yes, those German mystics like Bohr and Heisenberg have had their laughs about<br>
that up here... but not so many laughs... because, you see, they were dead wrong about<br>
something far more important to us, and we all can see it from up here.<br>
They were wrong about objective reality and the fact that time is not such a special dimension<br>
as THEIR theories made it out to be...<br><br>
Now since they themselves are up here, and cannot repent their sins directly to the world<br>
of earthly physics... it is my unfinished task to bring the truth to the cousins<br>
and descendants of Bohr... if there are any such here..."<br><br>
But then it would become mostly too technical, I suppose... though<br>
it would be nice if some of the new stuff could really be communicated...<br>
whatever... I had some thoughts about further words a week ago, but not now...<br><br>
<br>
==============<br><br>
No comments on instant evolution just now. <br><br>
One funny association, however...<br><br>
In the neural network field, one robotics guy was very proud how his neural net design could push <br>
an unstable cart around a figure-eight track.. and then, when the weight on the cart changed, adapt to<br>
near-perfect tracking again after three laps around the track. But -- in a book in 1990, I described<br>
how he could adapt much FASTER than three laps, if he used a Time-Lagged Recurrent Network<br>
(TLRN) instead of baby shrink-wrapped MatLab kinds of stuff. (Well, he was a little better than that...)<br>
And Ford later tried out the TLRN approach, and confirmed it does all I said.<br>
VERY fast adaptation, FOR environmental changes QUALITATIVELY SIMILAR to changes which<br>
were experienced before during the training of the network. Qualitatively similar does not mean<br>
a familiar POINT -- only a familiar SPACE. I suppose genetic systems might have<br>
some ways of implementing that kind of capability,, maybe..<br>
if they are that intelligent...<br><br>
Best,<br><br>
Paul<br><br>
<blockquote type=cite class=cite cite>
<dl>
<dd>
</dl></blockquote></body>
</html>