Discussion in 'Intelligence & Machines' started by Magical Realist, Apr 18, 2011.
Good, OK, fine.
Have it your way.
Log in or Sign up to hide all adverts.
I'm not going to debate with you or anyone else anymore, I will just ask you for your opinion without debating. What makes you think you're an simulation?
Do you really think life and human brain can be simulated?
How much computing power would you need?
I am talking about math, physics and the universe which I consider to be the question at hand. You appear to be talking about current computing power.
Read what I said, look at the links, see what I'm actually saying, and I'll be happy to discuss it with you. This forum is a discussion, not a debate.
Otherwise, we probably both have better things to do.
BTW, one of the "non sceintists" in my links was Alan Turrng. Remeber him?
But in answer to your questions, in hope they will be considred for a change:
I have never said I think we are living in a simulation. Where did you get that?
I think in the near future the human brain will be simulated by neural networks and or quantum computers. I can and will provide refrences if kindly requested.
I don't have that number at my finger tips. I have it in some of my books and will look it up and provide it "if" intellegent discussion starts up.
I will read those links, no doubt. But if you didn't say we're living in a simulation, did you mean that our universe is analogue to quantum computer or something else?
The following link is the response to Nick Bostrom's
I will respond to your last two questions when I get home. I need to consult my library.
The above is what I said. The no it's not is yours.
I direct your particular attention to: http://www.idsia.ch/~juergen/computeruniverse.html which I hope is of interest to you.
As to your answers to the last two questions, I offer the following:
Ray Kurzweil, in his 1999 book “The Age of Spiritual Machines” discusses duplication of the human brain in great detail. I suggest you read it, or other web based discussions concerning the subject.
Some of what Kurzweil predicts follows:
In his The Age of Spiritual Machines, written in 199, page 278, he predicts that in 2009“ A $1000 computing device (in 1999 dollars) will be approximately equal to the computing capacity of the human brain” Note: I paraphrase this and below so as not to be guilty of copyright infringement.
In 2029 he predicts that a $1000 unit will have he computation capacity of 1000 human brains.
He predicts that in 2029 that computers pass apparently valid forms to the Turing Test, but not necessarily in all of a humans diversity.
Page 280, for the year 2029, he predicts that machines will claim consciousness, and that this will be largely accepted.
As to my reference to Nick Bostrom's work, it is was taken seriously by various people at IBM during Brain Greene’s tenure there and is discussed in his book “The Hidden Reality” concerning the types of possible multiverses including the simulated and ultimate multiverses, chapter 10.
I still state, with flexible rigidly that the universe may be acting as a quantum computer.
Then again, it may all be an illusion. Mine and everyone else’s. Or just mine.
Ah no. In research for Strong AI, the interpretation on how a system becomes sentient requires it to work out it's own logic, a system that works for it. It does have governing parameters in it's initial programming but that's no different to having governing parameters in regards to your physical make up for you.
So a Strong AI could quite well define something that would be different from a Weak AI counterpart (which is programmed on the basis of decision matrix's already being pre-considered by the programmer)
As stated previously...
The universe shouldn't be assumed to be a Simulation but an Emulation. Simulations are used to make "false realities" that use "synthetic data". Emulations attempt to encompass absolute duplications, if not integrated fully as a/the "real".
Emulations seem to be more interesting hypothesis.
Could you be more specific, I have to admit I'm not familiar with emulation hypothesis, how would description look like. You said emulations attempt to encompass absolute duplications, does it mean they represent duplications of the original universe, by having the real data, unlike in simulation where is synthetic data?
What's difference? I thought both emulation and simulation are pretty much the same thing since they require computer to run them.
As far as I know (from wiki):
In computing, an emulator is hardware and/or software that duplicates (or emulates) the functions of a first computer system in a different second computer system, so that the behavior of the second system closely resembles the behavior of the first system. This focus on exact reproduction of external behavior is in contrast to some other forms of computer simulation, in which an abstract model of a system is being simulated. For example, a computer simulation of a hurricane or a chemical reaction is not emulation.
So does it mean emulation is not possible for real world reactions and evolutions like chemical biological, atmospheric, cosmic... and etc...?
How does this beat simulation argument?
It's not actually hard to simulate the universe and all of its atoms. The most we ever view at once is about 2000, take a look on the internet.. pictures of atoms.. the most you ever see is about 2000. The rest of the time you just see flat surfaces. Photons... no sign of them. Electrons... I have seen one on the internet, and it was a blur. With our current magnification we see very little. The brain.. we can think OK, but maybe we have a real brain when we are not in the simulator? So we can use that real brain.
Outside the simulator there can be a real world. We may be much more intelligent in the real world. Maybe this simulation only uses a part of our brain?
There are about 100,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000 atoms in the universe.
That's a few more than 2000.
Do you have any idea how you would go about imagining an electron, or an atom for that matter? I don't think you do. And what about photons?
You miss the point.. you only have to say the words..
for the simulator to work just as well as the real thing.
If you haven't seen the atoms, then you have only read the words, and so the simulator would only need to say the same words. This would be more like the Matrix movie where you don't need to test everything, just observe a simulated environment.
Hi, James, I'd like to ask you what is your opinion on this subject? As a physicist do you really think universe is an simulation of an quantum computer or is it something else? Or that universe is ANALOGUE to quantum computer?
I'm not here to debate anymore, because I feel it's somehow pointless, so I gave up.
Right now, I'm just asking other people's opinions and that's all.
Hardalee, I gave you the wrong link!
Read the following link:
This link is the respond of the Nick Bostrom's simulation argument.
My very bad mistake.
The ponit is no super-computer can match the plasticity and intelligence of an grown up human brain, that's the point you can't match the adaptability of human brain and instinct for survival, as well as the consciousness, AI robots are and alwys will be very limited in that aspect compared to human brain. Human brain is still a very huge mystery.
I agree with James R on this subject.
Dr. Bostrom’s theory does not accurately represent the differences between the computers we use and the brains which do our thinking. The fact is that, though our computers require many gigabytes to store a fairly low-quality video of a relatively short timespan, our brains are capable of storing vast amounts of memory which is not only visual and auditory, but which involves all of the senes. And we store this video constantly. Storage isn’t perfect forever, but the human brain clearly has orders of magnitude more processing power than do computers. The human brain is a computer to rival anything we will develop for probably the next thousand years, if we last that long.
Even assuming that civilizations other than ours eventually do build a super-bio-brain-computer with the capabilites Dr. Bostrom suggests, to waste such a marvel on running a simulation of other beings would be a crime. Furthermore, a simulation of other beings represents absolutely no benefit to any society or civilization. In fact, it is a waste of resources, completely, in any material sense. So in fact the likelihood of such simulations being run many times throughout the history of the universe is actually fairly slim. Furthermore, each day of our lives wherein we and our society and surroundings continue to exist provides further evidence that we are not, in fact, part of a computer simulation, because such a thing would surely not carry on for extended periods of time. For one thing, no computer can last indefinitely. Computers (of any kind) by their very nature must contain moving parts, which must eventually degrade and fail.
I must agree with Dr. Bostrom’s possibility #2, that simulations of the type mentioned are extremely unlikely, because simulating an entire civilization would require a significant concerted effort to design and maintain.
You really think there’s only ONE set of “simulators”? And that those simulators are not just another simulation, and so forth? Pretty narrow-minded thinking! You might want to ask the simulators simulators. Or what about their simulators?
Obviously, once a consciousness (doesn’t matter if it’s within an ID) reaches the point at which the simulation breaks down, the simulation/simulators would scramble to “patch the holes” so that the consciousness is “contained” within the “reality”.
The problem is, just as a hacker has an easy time once they find a flaw, it’s pretty easy once the initial break is made to just jump through all the levels of consciousness and come to the full realization that it’s all an illusion, and then you are just freed from the whole damn thing. Oops… game over.
It’s also kind of pointless to talk about computer processing power and focus on all this from the perspective of only the base senses that you think you have (touch, taste, smell, vision, hearing…).
Computers and technology have impossible time to even conceiving at least one of those senses.
Also, unfortunately this debate brings to mind attempts to cast doubt on real science with Intelligent Design theories. This “life is a simulation” theory is just a sci-fi version of creation myth. Dr. Bostrom’s peer at Oxford, Roger Penrose, has already discussed the impossibility of computing the universe. In a nutshell, the universe just isn’t big enough to simulate reality. Even the smallest microcosmos that contains living, thinking organisms would be impossible to simulate with the vastest, most powerful computers imaginable. All the atoms in the known universe would not be enough to build a computer that could properly simulate another universe.
All Bostrom provides is the point that if we are very redundant you can simulate the world/universe whatever numerous times over in a computer. It’s only really philosophically interesting if true and his argument is not convincing. Currently I can think of no examples of something that exists which we can accurately simulate in less space than the real thing except exceedingly trivial examples. Certainly nothing biological. Not even close.
Bostrom essentially states that neurons should be sufficient, but even if one needs to simulate some molecular aspects there’s probably a high level of redundancy and it shouldn’t increase the computational needs enough to kill his argument. No one has any idea if that’s true. All molecular geneticists say it’s extremely questionable. For instance learning quite likely triggers genetic circuits similar to those involved in development, so it must be molecular. Bostrom clearly knows too little to make estimates on the topic.
Taking such a rough grained approach to simulation could easily miss a lot of critical aspects of neuronal function which could easily put his calculations of computational cost off significantly enough that his whole argument is meaningless.
The same can be said about whether or not we can accurately estimate the capacity of nano-computers, maybe the current estimates are over estimates that assume certain computer designs work that don’t – people have barely made anything but the simplest nano devices and he’s talking about nano-computers. Furthermore, his cited estimates of computing power differ by 20 orders of magnitude which to me seems like a statement that people can’t accurately predict what they’re likely to be. He should make the argument only based on a reasonable extension of Moore’s law to currently agreed on physical limits. I mean furthermore he discounts cooling, which is certainly going to be an issue if you scale computing power of a sugar cube sized design to that of a planet like he does.
The hole here, is that somewhere, there has to be a real world that started the first simulation. If the people in the simulation ran a simulation from inside their own simulation, then ultimately that new simulation would be running on the original real computer. The one real computer would have to run everything, so the computational power may be far more than the authors’ are thinking.
I think this discussion is interesting, only because it demonstrates how contingent and changing epistemology is. As a historian, I see the computer as only the latest in a long string of technologies that has been used to define and shape the way we think about things we do not know, things like consciousness, the cosmos, and creators. We will always be looking to define the unknown using models and metaphors derived from that which we do, and such rhetoric is interesting, not so much because it is “the truth,” but because it is one barometer of our cultural moment, an index of the extent to which our own ways of thinking have been changed by some remarkably influential machines.
How we do know that we are sentient. At least, how do I know you are sentient? You really cannot answer any of this questions without defining setience and consciousness, which continues to elude.
Beyond this, however, I think the issue is what will become of humanity as our technological capabilities increaee: will we enter into a golden age of creativity, exploration and imagination? Will we degenerate into useless lumps who can nothing except entertainment ourselves, or will we just destroy ourselves before doing much of anything?
You guys are missing the point. What's of interest here is not the investigation of whether the universe is a simulation. What is interesting is whether the universe could be a simulation.
It is possible to prove that the universe is not a simulation. The absence of a Planck length or Planck time come to mind. In order for the universe to be computable (and thus possibly a simulation) there must be a upper bound on precision (assuming that the universe holding the simulation is in 3 dimensions of space and 1 of time).
Another interesting point is that the universe is not deterministic. This does not defeat the universe-as-simulation argument, but it makes it less plausible. A deterministic universe would be much easier to simulate than a non-deterministic one.
Separate names with a comma.