[ExI] Benchmarking the Singularity

Stuart LaForge avant at sollegro.com
Sun Jul 21 18:56:51 UTC 2019

> Hi Stuart, thanks for an absolutely first rate post, it was detailed yet
> clear. Really really good.

That's high praise coming from a skeptic like you so I appreciate it.

>> *The hilarious irony of the situation is that if my theory is correct,
>> then a human brain has to subconsciously perform tensor analysis in order
>> to reach the conclusion that it is lousy at math.*
> Damn, I wish I'd said that!

It's not too late. You can still say it. If the word gets out, then  
maybe math literacy will increase.

>> *> In other words, in terms of total number of neurons, the  human brain
>> is some 4 million times larger than AlphaGo's. In terms of  synapses it is
>> likewise on order 10^6 times smaller than the human  brain.*
> I doubt a computer would  need a million times more synapses to beat us at
> all intellectual tasks, for one thing the average informational signal in
> our brain moves about as fast as a car does on a turnpike while the
> informational signal in a computer moves at close to the speed of light.
> And I would bet money that the artificial neurons in AlphaGo's brain are
> organized in a more efficient less buggy way than the neurons in our brain
> are.

For playing go, perhaps . . . for figuring out new and creative ways  
to dodge predators while finding food and mates, probably not.  
Airplanes might fly better than birds, but they are not yet smarter.  
The human brain is not so much a single neural network as it is  
multiple interconnected neural networks built on top of one another.  
Like cities in Europe being built on top of ancient Roman forts.

> A raven's brain is only about 17 cubic centimeters, a chimpanzees brain is
> over 400, and yet a raven is about as smart as a chimp. And the African
> Grey Parrot has demonstrated an understanding of human language at least as
> deep as that of a chimpanzee and probably deeper, this despite the fact
> that the chimp's brain is about 25 times as large.

Yes. Birds in general are marvels of evolution. Ravens and Parrot-kind  
including the kia, have some of the densest brains known. They have as  
many neurons in their much smaller skulls as ungulates like horses do.  
I imagine the trade off was fewer glial cells or something.

> I suppose that when
> there was evolutionary pressure to become smarter a flying creature
> couldn't just develop a bigger, heavier more energy hogging brain; instead
> of the brute force approach it had to organize the small light brain it
> already had in more efficient ways.

The adaptations of birds are incredible, especially their extended  
respiratory system. The air sacs in their hollow bones operate like a  
second set of lungs. Every time a bird exhales, it is exhaling air it  
inhaled 2-3 breaths earlier and every last bit of oxygen has been  
pulled out of it.

> Our brains are about 1400 cm, but I'll
> bet centimeter by centimeter ravens are smarter than we are. Being called a
> birdbrain may not be an insult after all. For this reason I believe if one
> wishes to study the nature of intelligence then crows and ravens would be
> ideal candidates, compared with other animals their brains would be more
> elegantly designed and have less spaghetti code and hard to understand
> kludges.

Very possibly.

Stuart LaForge

More information about the extropy-chat mailing list