Slashdot Mirror
Artificial Neural Networks Demonstrate the Evolution of Human Intelligence
samazon writes "Ph.D. students at Trinity College in Dublin have constructed an artificial neural network model to demonstrate the Machiavellian intelligence theory — that human intelligence evolved based on the need for social teamwork and indexing a variety of social relationships and statuses. (Abstract) The experiment involved programming a base group of 50 simulated 'brains' which were required to participate one of two classical game theory dilemmas — the Prisoner's Dilemma or the Snowdrift game. Upon completion of either game, each 'brain' produced 'offspring' asexually, with 'brains' that made more advantageous choices during the games programmed to have a better chance to reproduce. A potential random mutation during each generation changed the 'brain's structure, number of neurons, or the strengths of the connections between those neurons,' simulating the evolution of the social brain. After 50,000 generations, the model showed that as cooperation increased, so did the intelligence of the programmed brains."
The full paper is available.
Now *THAT's* intelligent design!
$Ducks
Have gnu, will travel.
Intelligence is really a socially directed kd-tree...
They had a whopping 20 neurons (nodes).
Wouldn't this be more like a model of insect intelligence, say from about 250 million years ago ? Maybe it could explain the evolution of bees.
The paper suggests that evolution favors cooperation but that it also favors low-cost solutions (i.e. lots of little dumb brains (ants) vs. singular powerful brains (humans)). Perhaps this explains the Fermi Paradox: Aliens are all over the place on other worlds, but they're mostly the former kind of cooperative rather than the latter.
Great warrior...hrmph! Wars not make one great.
So if we stipulate in our environment that smarter brains are more likely to reproduce, then the smarter brains reproduce just like would happen if human brains evolved to be smarter as a competitive advantage, so human brains evolved as a competitive advantage? They've stacked the dice to make evolution happen in their artificial world, so why should we make the inference that the world's dice are stacked in just the same way?
I've seen "Idiocracy". This can't be true.
...richie - It is a good day to code.
So *someone* creates the neurons, creates connections, establishes rules of selection, creates mutations for random change, turns the system loose, and that is somehow a model for completely random unguided uncreated evolution???
So did they evolve the optimal strategy of starting with cooperation then mirroring your opponent's last move? Because it's cooperative but you don't need many neurons for that.
If this reasearch is correct, people living under Socialism (where cooperation is high because it is mandatory and forms the very basis of the system) would wind up becoming smarter than people who live under Capitalism (where competition/versus behaviour and everyone-for-himself thinking is closer to being the norm, relatively speaking). This is an interesting result from a politics perspective, because proponents of unfettered, hard-core capitalism often charicature socialist systems/ideologies and the people who live under them as being "unfree" or "living like cattle". Quite a contradiction of views/results, eh? One could infer from this, that hard-core capitalists lack the "cooperative intelligence" that a working socialist system creates to such an extent, that hard-core capitalists cannot even comprehend how a cooperation-socialism based system works - because they never developed a comparable cooperation-based intelligence themselves. Of course it is somewhat silly to arrive at this result purely based on a clever neural-networks exercise that has been conducted by some PhD students. But the hypothesis of capitalists never developing, and thus largely lacking, the cooperative-intelligence needed to live in a socialist society is interesting nevertheless, is it not? It might explain why hard-core capitalists are so keen at swining rhetorical wrecking balls at anything that has even a small whiff of "socialist cooperative model" to it... Like Open Source Software - which is very much a product of cooperative intelligence - for example. To put it more simply, maybe prominent capitalists like Bill Gates cannot help dissing open source efforts like Linux, because they are, from a personal development standpoint, largely unfamiliar with the kind of cooperative-intelligence that a more "socialist" mode of product creation promotes.
Why did the chicken cross the road? Because Elon Musk put an AI chip in its head.
Now that's intelligent ;)
Does this formulier work?
i like it
My CPU is a neural net processor - a learning computer
Fact: The key to any successful cooperative test is trust, and as our data clearly shows, humans cannot be trusted. The solution: robots! Then, fire the guys who made those robots, and build better robots. Then, run those robots through a regimen of trust exercises, creating a foundation of mutual respect, reinforced by the simulated bonds of artificial friendship. Inspiring stuff. And finally, we put that trust to the test. Bam! Robots gave us six extra seconds of cooperation. Good job, robots. Cave Johnson. We're done here.
https://www.youtube.com/watch?v=AZMSAzZ76EU
Upon completion of either game, each 'brain' produced 'offspring' asexually
You don't say...
The title made it sound like this was something big. In reality it's interesting but not that interesting. I've see this kind of thing before when they did something similar to get a "team" of robots to play a game together. Nice, but not human level intelligence. Heck it's not even cat or dog level intelligence. I was hoping this would be some kind of break though AI but really it's more disappointing reading the article.
Not Applicable in Tennessee
With over 7 billion people working together... just think where we could go with intelligence. But we have ignorant politician and commanders in the way.... Why is that?
... is the same problem pretty much every study has -- it's based on the concept of ceteris paribus, which does not exist in reality.
From a purely academic standpoint, however, it is a neat experiment.
An enigma, wrapped in a riddle, shrouded in bacon and cheese
Ok, we're not gonna get Jude Law Sex Bot (not related to SpongeBob SquarePants) out of this AI experiment, but it seems nifty. I wrote something similar to test clonal interference. Worked nicely, too. It's easy to enforce evolution on your models. It's when they do something unexpected that the fun begins. Add enough computing power, and you can increase the latitude of the parameters you're testing. Hey, better yet, make the whole thing tiny, complex, and chemical in nature. Coil it up! Wrap it in a factory with little thingies that work like code interpreters. Yeah! That's the ticket! We can call the process Learning Intuitively For Experience, or LIFE, for short. Not that it'll catch on.
What about thinking? Hmm. Well, 100 1st graders can be quite innovative if you lock them in the Lego store at night. Every morning, remove the ones who used blue pieces in their constructions. Why? It's MY experiment!!! If you can't get 1st graders, then alpacas or fire ants will do. Eventually, you will have a most excellent Lego creation. Ok. Will you understand what it is, though? Maybe it makes sense to them, even if we don't see the utility of it. Maybe it's their "art"?
So now you see why this is all interesting, but not terribly important... ...yet.
LOL!
Is a loose demonstration of plausibility of the Mach. theory.
-- The Grand Teddy Bear has Spoken: "Windows 8 Source Code Available NOW! more disgusting than your pr..."
Nothing would exist (big bang could not have even occured) without the existence of the laws and parameters that govern time, space, matter, energy, physics and chemistry. This lead to the undeniable conclusion that an utimate Intelligence was created simultainously with the big bang. This does NOT lead to the conclusion that this Intellegence has taken control in any aspects of the 13.7 billion year evolution of the expanding universe. It has only created the tools with which natural processes have to work; from the creation of most of the elements in the periodic table to the functioning of the human brain.
Great idea to use a neural network to model intelligence in this context, but this has nothing to do with Darwinian evolution, as several posters seem to think. Their GA mutated network weights and nodes. The weights would be equivalent to neural synapse strength in the brain. The nodes would represent blocks of cortical neurons that span many levels. In fact they could represent entire neural subsystems.
According to the theory of Darwinian evolution, the entire brain developed - as did everything about us - through natural selection of accumulated random mutations of individual genes. Even the most simple mutation in this study - that of network weight - would likely require the beneficial mutation of many genes. If all relevant genes did not mutate simutaneously, mutation in any subset would be unlikely to be beneficial. For even a few genes to simultaneously mutate beneficially, the probability of occurence is so high as to be basically impossible (see the history of the arms race between humans and plasmodium malariae). Now consider the sudden addition or removal of one of the authors' nodes, which would likely require of simultaneous mutations. Yeah, not gonna happen...
Republicanism is against collaboration and group work as well as intellectual elites. So they are bringing us down since they try to move us away from the process that promotes the evolution of human intelligence?
My apologies to any republicans reading this, I do not mean to hurt your feelings by pointing out your low intelligence or imply you are destroying mankind especially when you are incapable of taking responsibility for your actions.
The system is set up such that we can only vote them in, and not fire them.