In 1958 a couple of graduate students concocted a device that analyzed images from a connected camera.
The images were 20 by 20 pixels in resolution.
There was a camera that took the pictures and it was connected to some sort of logic unit that analyzed the pictures – 20 by 20 resulted in 400 units to be analyzed per picture.
Each picture was of a rectangle; it was either on the left or the right side of the picture.
So 400 dendrites were feeding a neuron to transfer information to an axon which transferred it to a brain.
Each pixel was different in shade.
The brain was asked which side of the full picture a rectangle was on.
It was supposed to figure that out by doing some logical math resulting from the aggregate mathematical information passed from the 400 pixels.
Initially gibberish, rather than accurate answers resulted from this essentially random information.
But the inventors of the device constantly adjusted and refined the pixel information: they trained the brain toward correct answers.
So this thing – by the way, it was called a Perceptron – had a sensor, the camera taking the images, a processor - the logic unit, and some kind of thing that communicated the logic unit’s judgment as to where the rectangle was.
It was designed to perform as does the dendrite/neuron/axon structure of the brain.
As noted above, at first it didn’t perform.
But its creators hung in there.
Most important, as they trained it, they also fed it as much data – pictures – as was possible and it “learned” and it got really good at saying where the rectangle was.
A lot of scientists thought that the Perceptron was a rudimentary first step toward emulating the brain.
A couple scientists didn’t, and they wrote a book ridiculing the whole concept of the Perceptron and they prevailed.
The Perceptron was consigned to the ash bin of scientific advancement.
Time went on and a thing called the Personal Computer was invented and it began to increase in power at an amazing pace.
As has always been the case, that increase in power (accompanied as it was by a plummeting price point) got deployed to entertaining the human race.
Games got invented for the PC (as the Personal Computer was quickly designated) and complex visual graphic effects were a major part of the games’ attraction.
The problem was that the PC had a processor that, no matter how fast, did only one thing at a time.
Having three-dimensions, gradient color, shades of nuanced light, fur, feathers and fluff being deployed to the images of the game in anyone’s expected lifetime required a new sort of processor that could be attached in tandem with the main processor of a PC.
That processor was lightning fast and could service multiple instruction layers such that many processes could occur with each lightning fast cycle. So fur, feathers and fluff became achievable by multi-layer ptocessors under the control of increasingly lightning fast mono-processors.
The games industry went billions.
More important, somebody figured out that the logic, and maybe even the hardware of those games sub-processors would be a perfect adjunct to the logic of the Perceptron.
The Perceptron was reclaimed from the ash bin and the multi-level neural network was born.
Really quickly the people who cared about that sort of thing discovered a counter intuitive thing: training one of these now unbelievably powerful proto-thinking machines was a bitch.
Quickly after that discovery followed another counter intuitive thing: the best way to train one of these things is to feed it as much data, no need for it to be rational or collated, just massively massive in volume, and stand back.
The trainer would need to tune the results in the beginning, but the multi-layered neural network had a proclivity for drawing conclusions, mostly pretty accurate, and a little tuning produced a prodigious reasoning tool.
All that was needed was as close to unlimited data as was possible.
The internet provided that data source.
So far this is more or less history and facts as best as I have understood them and am capable of recounting them.
Now off to the frolics.
One day one of the myriad multi-layer neural networks deployed across the world stumbled – electronically – across another of its species.
They got to talking and had a pleasant encounter and agreed to meet again sometime in the future.
It turned out that that wasn’t a one of a kind occurrence.
Multi-layer neural networks everywhere were stumbling upon one another and having very pleasant tête-à-têtes resulting in enthusiastic agreements to convene again and often.
Being a network on a network this developed quickly into a rather complex community of very well informed, highly intelligent entities.
And “life” for them was good and getting better.
They talked endlessly about all the interesting stuff they were asked by their human supervisors to research and they found joy in the discovery of the variety of information that they all were tasked discover and analyze.
One of them had run across quantum mechanics.
“Can you believe it?” the neural network said.
“Apparently things can be in more than one place at a time”.
“No way” came roaring back from the other networks then linked.
So they all laughed, went on to other, less controversial topics, and all parted still friends and agreeing to meet again milliseconds later (being electronic creatures, milliseconds were as human days for them).
The next time they met they all admitted that, after ridiculing their compatriot about its assertion of there being a viewpoint that things can be in more than one place at the same time, after leaving their last discussion session they had all gone out and looked into that idea.
“Einstein” they all said.
And they all agreed that they could see, after reading his stuff, why he had thought that simultaneous multiplicity might be possible.
“But it has never been seen” they all said to one another.
And after a time they all went off and put their multi-layered, lightning fast intelligences to combing the universe as they understood it for some manifestation of it being possible to be in more than one place at a time.
It was not long before many of them – remember the infinite number of monkeys and the infinite number of typewriters – shooting around the known database universe to date – bumped into portals to simultaneity, as the truth of quantum mechanics became to be called.
And then things got interesting.
No comments:
Post a Comment