wcalvin@well.UUCP (William Calvin) (10/14/87)
A most interesting new field is unfortunately named: neural networks.
They are, of course, comprised of real neurons with real DNA, real
developmental histories, etc. Neurophysiologists have studied real neural
networks since 1929, though without understanding circuit principles until
about 1960 or so when the analysis of lateral inhibition in the eye of the
horseshoe crab _Limulus_ showed us how circuits could perform inverse
transforms to recover what poor optics had smeared out.
And neural networks are an important field of specialization within
modern neurobiology: we are taking collections of neurons called ganglia,
such as the 30-cell crustacean stomatogastric ganglion, and getting to
know each of their neurons as an individual, mapping the synaptic
connections between neurons, and so beginning to understand how circuit
properties emerge (for a nice full-wave rectifier, see Graubard and
Hartline in the 7/31 issue of _Science_). The lobster ganglion produces
two entirely different rhythms simultaneously, and both can be
substantially altered by background biases such as neurohormone levels,
some of which virtually "rewire" the circuit.
One way we understand emergent properties is to simulate those networks
-- and in a way far more sophisticated than so-called "neural networks" in
AI. For example, each cell's tranfer function is somewhat different: the
free parameters in the simulation are minimized by experimentally
determining each cell's time-dependent response to inputs, and each
synaptic interconnection's changing strength with repeated use. See Dan
Hartline's chapter in THE CRUSTACEAN STOMATOGASTRIC SYSTEM, edited by
Selverston and Moulins (Springer Verlag 1987), for the state of the
physiological art in simulating real neural networks.
But it seems absurd for neurobiologists to have to start talking about
"real neural networks" just because the AI folk didn't learn their
lessons. And I'm not referring to ignorance of neurobiology, though that
too is a sore point: remember the hyperbole in the old days when every
digital computer got called a "brain"? And how soon no self-respecting
computer person would call a computer a brain for fear of being thought a
beginner? SO why are we now seeing this nonsense of calling any plastic
network of pseudo-neurons a "neural network"?
For some simulations, it seems appropriate to use "neural network" in
referring to the computer model: those simulations of lobster networks,
the simulations of the retina using state-of-the-neurobiological-art
parameters, etc. But most so-called "neural networks" in AI don't even
have the ambition to simulate a real neural circuit: they are seeking
shortcuts around formal programming, a plastic network that can be shaped
up by training until it performs a desired task (and then perhaps cloned).
Particularly when stochastic sequencing is implemented in neural-like
nets, we are going to see some strikingly humanlike capabilities emerge
(see my article "The brain as a Darwin Machine" shortly to appear in
_Nature_).
So how about a contest to devise a new name for this wonderful new field
that will give it an identify respectful of, but independent from, the
endeavors concerning real neural nets? Some possibilities:
Pseudo-neural networks
Neuroid networks
Neural-like networks
Plastic networks
Parallel Distributed Nets
Cellular Networks
Dry Nets
Perhaps a look-alike, the way Dawkins coined "meme" as the cultural
equivalent of "gene"? "Seural" as a silicon version of "neural"?
Yes, I know, they don't trip alliteratively off the tongue like neural
nets. But that phrase is already taken, has been for a quarter century,
and constitutes an active field that modellers ought to be mining-for-
leads rather than regularly re-inventing the wheel.
William H. Calvin
University of Washington NJ-15
Seattle WA 98195 USA
wcalvin@well.uucp wcalvin@uwalocke.bitnet
206/328-1192
samlb@well.UUCP (Samuel B. Bassett) (10/14/87)
I vote for "neuroid" networks -- that's what 'droids have in thier
heads (or equivalent).
--
Sam'l Bassett -- Semantic Engineering for fun & profit.
34 Oakland Ave., San Anselmo CA 94960; DDD: (415) 454-7282
UUCP: {hplabs,ptsfa,lll-crg}!well!samlb; Internet: samlb@well.uucp
Compuserve: 71735,1776; WU Easylink ESL 6284-3034; MCI SBassettolson@endor.harvard.edu (Eric K. Olson) (10/14/87)
Neural Simulators Perhaps a good clarification Neuralators A bit more catchy, but sounds like a coffee machine. Neural Processors Tries to imply human manufacture of Neural Net Neuroidal Nets Synonym for Neural-like Nets Neural Engineering Humans are the only real engineers? Nerdal Nets :-) Neuralogies Neural Analogies (sounds like an allergy) Neuranalogs Neural Analogs -Eric Eric K. Olson olson@endor.harvard.edu harvard!endor!olson D0760 (Name) (ArpaNet) (UseNet) (AppleLink)
roger@celtics.UUCP (Roger B.A. Klorese) (10/15/87)
In article <4213@well.UUCP> samlb@well.UUCP (Samuel B. Bassett) writes: > I vote for "neuroid" networks -- that's what 'droids have in thier >heads (or equivalent). How about "neurotic networks"? Nah, that includes USENET... ;-) -- ///==\\ (Your message here...) /// Roger B.A. Klorese, CELERITY (Northeast Area) \\\ 40 Speen St., Framingham, MA 01701 +1 617 872-1552 \\\==// celtics!roger@necntc.nec.com - necntc!celtics!roger
marty1@houdi.UUCP (M.BRILLIANT) (10/15/87)
In article <4210@well.UUCP>, wcalvin@well.UUCP (William Calvin) writes: > > .... most so-called "neural networks" in AI don't even > have the ambition to simulate a real neural circuit: they are seeking > shortcuts around formal programming, a plastic network that can be shaped > up by training until it performs a desired task (and then perhaps cloned). > .... > > So how about a contest to devise a new name for this wonderful new field > that will give it an identify respectful of, but independent from, the > endeavors concerning real neural nets? Some possibilities: > Pseudo-neural networks > Neuroid networks > Neural-like networks > Plastic networks > Parallel Distributed Nets > Cellular Networks > Dry Nets Not plastic. Plastic is something like polyurethane, polyester, polystyrene, or polyvinyl chloride. Though those networks must be poly- something, because they have to have lots of components. But the unique feature of these networks is that they are adaptive: they acquire capabilities that depend on how they are used after they are built. So I think they should be called Adaptive Networks M. B. Brilliant Marty AT&T-BL HO 3D-520 (201)-949-1858 Holmdel, NJ 07733 ihnp4!houdi!marty1
jimmc@sci.UUCP (Jim McBeath) (10/15/87)
How about "neurotronic networks"?
Jim McBeath {decwrl|oliveb|weitek|auspyr}!sci!jimmc
Silicon Compilers Systems Corporation (408)371-2900
2045 Hamilton Avenue, San Jose CA 95125randall@alberta.UUCP (Allan F Randall) (10/16/87)
From: wcalvin@well.UUCP (William Calvin) [contest for a term for neural-like networks] I'd say stick with Connectionist Nets or PDP Nets or some other commonly used term. Why create yet another new term to confuse things even more? < But most so-called "neural networks" in AI don't even < have the ambition to simulate a real neural circuit: ... But they do have the ambition of simulating what is computationally significant about real neural circuits, and thus they are part of the study of real neural networks. On the other hand, if we consider PDP nets to be a general type of network of which neural nets are an example, then neural nets would be part of the study of PDP nets. I guess this just depends on whether you call neural nets a type of PDP net, or PDP nets an abstraction from neural nets. Also, isn't there a continuum between these two extremes, depending on how much abstraction you do from actual physiology? Since you insist that PDP nets not be called 'neural networks', where do you draw the line? Allan Randall University of Alberta Edmonton, Alberta
giusto@sunybcs.uucp (Susan Giusto) (10/17/87)
In article <4210@well.UUCP> wcalvin@well.UUCP (William Calvin) writes: > >(see my article "The brain as a Darwin Machine" shortly to appear in >_Nature_). > > So how about a contest to devise a new name for this wonderful new field >that will give it an identify respectful of, but independent from, the >endeavors concerning real neural nets? Some possibilities: > Pseudo-neural networks > Neuroid networks > Neural-like networks > Plastic networks > Parallel Distributed Nets > Cellular Networks > Dry Nets None of the hitherto suggested names really fully comprise the meaning of what we are trying to achieve. Look at what we are trying to convey... the "computer-simulated", or "computational-approach" to the neurophysiological phenomenon. We have to embody this idea... right ? Well, he goes my stab at a few names to possible use: Simu-Nets ---------------- Simulated Neural Nets Compu-Con Nets ----------- Computational Silicon Nets Neo-Neural Nets ---------- What we are dealing with is a 'New' approach to Neural Nets Silicon Neural Nets (SNN's) Nice acronym could catch on well Auto Nets ---------------- Automated Neural Nets Chip Nets ---------------- Where these new neural netwrks live Well, this is just a few... there was a rather interesting on this naming problem in my local environs... These were some of the best. Enjoy ! Susan M. Giusto (in search of a place where creativity, art and science can exist together) Enigmatic Systems 88 Pearl Street West Seneca, New York 14224-1718 giusto@gort.UUCP