I like to drive at night. I drive like a rat in a maze where I twist in circles and follow a complex pattern that I have memorized so well I could probably do it blindfolded. (That is, if rats could drive a crappy rust bucket that is leaking oil)
In any case, while I mindlessly drive the maze of streets that I’ve memorized my eyes have a tendency to catch whatever small changes that have occurred on the various routes. It’s sort of like flipping through a photo album except that you spot the difference instead of reminisce.
My mind also wanders and I find myself in crazy man debate. The kind of debate where crazy man goes into the hermit hill and bounces ideas back and forth. And this nonsense eventually led to computers and artificial intelligence.
Now I do not program artificial intelligence so I haven’t the faintest idea of how it is done. In any case, it was that mental tennis game that made me search for the following Ted Talk below.
Date Filmed: May 2007
Length: 20 minutes, 16 seconds
Total Views so far: 602,588
One Sentence Summary: “Human brain theory is revolves around prediction and notice how this doesn’t involve computers.”
PS> “Computers which can envision the future from memory and sensors are the future.”
Note: Hawkins speaks so much about brain theory that computers don’t really come into play until the very very end.
Now if computers ever came to the point that they would be able to comfortably program themselves via sensory data, I wondered if what would the best way to achieve this to be?
Using the human brain as the sample, what if it was two computers that were designed to program the other, and that they were specific to what details they could program?
For instance, left and right brain control logic and creativity. So what if, one side of this AI was looking out for ‘creative’ inputs (what’s new, creating identifications) and the other side was strictly looking out for ‘logical’ inputs (mathematical identifications, examining trial and error).
And the piece where I became stuck, was that the logical side of the brain is set up to be programming the creative side while the creative side is programming the logical side.
Like I said, I’m not out to program AI. The closest I have ever come to viewing the process was watching a student attempt to program a chess game.
In any case, Hawkins brings a very fascinating and amusing talk on how the brain functions and the perceptions of how it works.
‘We’re brains talking to other brains’, is a good quote, because it combines well with another good quote he makes which is (paraphrasing) ‘if you see a nose where you’re predicting an eye you go Holy Shit!’.
So what happens when a brain attempts to talk to a nose? It’s a Holy Shit moment that fires off in the brain and suddenly you’re on tilt of the situation.
Doesn’t this describe every argument, misunderstanding or conflict? You’re a brain trying to talk to another brain and you end up bumping into a nose. That’s like, what the fuck is your nose doing where your brain should be? What are you some kind of nose brain monster?
This kind of talk is more easily understood by replacing ‘nose’ with something more survival instinct like ‘stomach’ or ‘penis’, but you get the idea.
What’s really interesting is the fact that this talk is from 2007. And since it’s five years later, there has been a hell of a lot of advancements in computers (and treos) since that time. For one, a month after this Tedtalk, the iPhone was released to the unsuspecting public and that probably changed a whole mess of things. (When it comes to handheld devices)
What caught me off guard
Wow, if I was blindfolded I would have thought this was a younger Clifford Stoll speaking. Seriously.
This guy is all over the place, and he’s making jokes in the forms of statements that I’m not sure if he seems them as jokes. It’s really amusing nonetheless.
And Hawkins brings up some very interesting things about the nature of brain science. Instead of breaking the brain up into particular parts (say, in the way Jill Taylor did) it’s more about the entire experience. (Also reminded me of Future Physicist work in a way, but I digress)
Human memory works by first processing the data. If you’re in the dark and something walks by, you won’t remember it because it was dark. Same applies if it’s a piece of logic you’re unfamiliar with or something is truly ‘new’ while you’re new to it all. All that’s wiped out and ignored like a mathematical proof defining the initial assumptions and variables.
I think Hawkins said a few things that were just outright incorrect, but it doesn’t matter if they are or are not incorrect because that result doesn’t apply to his talk. It’s a very high level mode of thought that few people take the road on because it invites contradiction.
However, my biggest surprise is the fact that this Talk included the term ‘computing’. Hawkins really doesn’t drill deep enough into computers to explain how a computer will sense or what it means to truly ‘remember’ or have a memory of something. On the other hand, learning how the brain works is really quite rewarding.
It does make me wonder about the singularity idea. The idea that eventually computers will become so ‘intelligent’ that they are able to upgrade themselves faster than humans are capable and at a speed that is faster and faster until eventually computers simply handle everything.
That’s my description, I can probably find a better one.
The Singularity: http://en.wikipedia.org/wiki/Technological_singularity
There’s bound to be a few people who might want to argue that the Singularity is the future. That it will be mankind’s future or the computer’s eventual future or maybe it’s for computers that drive through mazes like rats. I don’t really know.
What I do know is that every once in a while, I go ‘Holy Shit, that’s a nose!’.
– Updated every Friday