help with philosophy...

[mofo]

to explain how a computer works our professor spoke about the chinese room.

in this situation there is a man who speaks no chinese and has never seen it, in a room with an "in" and an "out" bin similar to "input" and "output" on a computer. also in the room is a chinese-to-chinese handbook that basically tells you what to respond with if a certain sequence of characters are inputted. the information coming in is all chinese. the person goes to the handbook and finds the incoming sequence and then outputs a sequence associated with it. overtime the person would get quicker at outputting only because he recognizes the characters, not because he understands them.

this example shows how computers can only do simple character exchange. tying this back to the original question posted in the first post, there is no way you could "teach" a computer to feel frustration because they won't actually feel it the same way a human does.

the argument that this particular philosopher (Searle) makes is that you can only make a strong AI if you make it out of the same stuff as humans, implying that only clones of humans will ever be true strong AI.

[liftsiron]

A computer will perform as programmed.

[decadecadeca]

Do some reading on Gödel Machines....

"The growing literature on consciousness does not provide a formal demonstration of the usefulness of consciousness. Here we point out that the recently formulated Gödel machines may provide just such a technical justification. They are the first mathematically rigorous, general, fully self-referential, self-improving, optimally efficient problem solvers, conscious or self-aware in the sense that their entire behavior is open to introspection, and modifiable. A Gödel machine is a computer that rewrites any part of its own initial code as soon as it finds a proof that the rewrite is useful, where the problem-dependent utility function, the hardware, and the entire initial code are described by axioms encoded in an initial asymptotically optimal proof searcher which is also part of the initial code. This type of total self-reference is precisely the reason for the Gödel machines optimality as a general problem solver: any self-rewrite is globally optimal—no local maxima!—since the code first had to prove that it is not useful to continue the proof search for alternative self-rewrites."

[decadecadeca]

again, we program ourselves. we're fully capable of evolving on our own without having to do routine maintenance. with computers they can only really do so much as you tell them to. you can't take a computer programmed to work in a car factory to work in a wheat plant and expect it to adapt. it's programmed to do what it was intended to do.

the example the professor used in class was C3P0 and R2D2 and how they were the ideal example of "strong AI". they were allowed to move around on their own and had a sense of "Free Will".

I think you're mixing "knowledge" and "learning" just a bit.

An infant kept completely alone in a sterile environment will never program themselves. They will be totally nonfunctional for all intents and purposes.

And you're limiting your argument to current technology. Yes a specific robot will not adapt to a variety of work at this point. In a few years though a worker class machine could be constructed that wasn't task specific. Your question was what could be, not what is.

[mofo]

I disagree.

We're born with a genetic code, a set of programmed instructions. Over the course of our lives the people we interact with input data and we process it, learn from it and adapt.

No different than a computer. The more a person learns the more options they have in their programming menu, again no different than a computer.

again, we program ourselves. we're fully capable of evolving on our own without having to do routine maintenance. with computers they can only really do so much as you tell them to. you can't take a computer programmed to work in a car factory to work in a wheat plant and expect it to adapt. it's programmed to do what it was intended to do.

the example the professor used in class was C3P0 and R2D2 and how they were the ideal example of "strong AI". they were allowed to move around on their own and had a sense of "Free Will".

[decadecadeca]

it's more of an evolutionary thing in humans and more of a technical thing in computers. a person doesn't go around to babies and tweaking their brains. and even then, in humans, the learning gets more and more technical so even the learning evolves in humans more than it does in computers.

I disagree.

We're born with a genetic code, a set of programmed instructions. Over the course of our lives the people we interact with input data and we process it, learn from it and adapt.

No different than a computer. The more a person learns the more options they have in their programming menu, again no different than a computer.

[mofo]

Aren't humans programmed to learn?

it's more of an evolutionary thing in humans and more of a technical thing in computers. a person doesn't go around to babies and tweaking their brains. and even then, in humans, the learning gets more and more technical so even the learning evolves in humans more than it does in computers.

[the art of war]

^ I'd love to believe that.

[decadecadeca]

good article. we don't need citations. this is an informal paper. we simply need to fight for out point of view, then play devil's advocate and then conclude with why the counter argument is wrong. he jokingly said that if we use ANY citations that we'd fail the paper.

reading through your part of the argument, aka the land-navigating robots, you could argue that they were programmed to do so. that strong AI would be able to be placed on a planet with the simple directions of navigating. the robots in your example have been programmed with learning algorithms and such so they learn only because they have already been programmed to learn and told to not take routes that are difficult to get through.

Aren't humans programmed to learn?

[mofo]

It depends how you define experience. If you mean by the fact is can happen, the computer will experience frustration, just not in the sense that a person would. The computer may play a game, and realize it is not winning. If it then turns itself off, that wold be a form of frustration.

As for the materialist, chemicals are king. New Love is the release of hormones in the brain. Once these chemicals fade, what remains is true lov.e If you keep pumping someone full of these chemcials, the NEw Love won't fade. Therefore, you can make someone fall into love.

In order to make them fall out, if you can find how to elimiante these chemicals, you would force someone to fall out of love, and possible lose all happiness

but who is to say everyone experiences any emotions as the next person. for example, if you take antisocial personalities and look at their brain scans you'd see their pleasure is rooted in different places than the average person's and they also seek pleasure from greatly different things.

[mofo]

What level of "understanding" is that?

Computers can already learn from past mistakes. Land navigating robots placed in a maze will learn from blocked routes and not take that path again, essentially "learning" how to successfully exit the maze.

"We have shown that evolution of learning-like properties is possible without
modifications of synapse strengths, but simply by relying on complex internal
dynamics of CTRNNs........However
neurophysiological experiments have indicated that the way animals and
humans perceive, classify, and memorize, for example in the olfactory system, is
by transitions between chaotic attractors in dynamical systems formed by large
numbers of neurons in the brain. These results correspond nicely with the
view of memory and learning presented in this paper."

http://infoscience.epfl.ch/record/63...l_evorob03.pdf

I'd cite it properly but I don't know if you're using APA or MLA.

good article. we don't need citations. this is an informal paper. we simply need to fight for out point of view, then play devil's advocate and then conclude with why the counter argument is wrong. he jokingly said that if we use ANY citations that we'd fail the paper.

reading through your part of the argument, aka the land-navigating robots, you could argue that they were programmed to do so. that strong AI would be able to be placed on a planet with the simple directions of navigating. the robots in your example have been programmed with learning algorithms and such so they learn only because they have already been programmed to learn and told to not take routes that are difficult to get through.

[decadecadeca]

we're talking about AI and how we will never build a "strong AI" because there is a level of "understanding" that humans experience that AI cannot. .

What level of "understanding" is that?

Computers can already learn from past mistakes. Land navigating robots placed in a maze will learn from blocked routes and not take that path again, essentially "learning" how to successfully exit the maze.

"We have shown that evolution of learning-like properties is possible without
modifications of synapse strengths, but simply by relying on complex internal
dynamics of CTRNNs........However
neurophysiological experiments have indicated that the way animals and
humans perceive, classify, and memorize, for example in the olfactory system, is
by transitions between chaotic attractors in dynamical systems formed by large
numbers of neurons in the brain. These results correspond nicely with the
view of memory and learning presented in this paper."

http://infoscience.epfl.ch/record/63...l_evorob03.pdf

I'd cite it properly but I don't know if you're using APA or MLA.

[Konitz]

I say yes it is possible that someday a computer will reach a level near human consciousness. Animals like cats and dogs clearly feel frustration, they become anxious, scared and nervous as well. Even mice populations have demonstrated levels of frustration when increased numbers are kept in close proximity.

If relatively "simple" creatures can display those emotional responses I don't think it's completely unreasonable to think a computer will be able to do the same at some point in the future.

The tipping point will be a computer that can demonstrate true free will and not a basic formulaic response from programming.

Watched a show on how they took brain matter and hooked electrodes up to it it started communicating with each other; they did not have the technology to understand what it was saying but eventually I think there will be a time when they can take brain matter and put it into a robot application. I guess you would call that a cyborg.

[decadecadeca]

I say yes it is possible that someday a computer will reach a level near human consciousness. Animals like cats and dogs clearly feel frustration, they become anxious, scared and nervous as well. Even mice populations have demonstrated levels of frustration when increased numbers are kept in close proximity.

If relatively "simple" creatures can display those emotional responses I don't think it's completely unreasonable to think a computer will be able to do the same at some point in the future.

The tipping point will be a computer that can demonstrate true free will and not a basic formulaic response from programming.

[alwaysgrowing]

It depends how you define experience. If you mean by the fact is can happen, the computer will experience frustration, just not in the sense that a person would. The computer may play a game, and realize it is not winning. If it then turns itself off, that wold be a form of frustration.

As for the materialist, chemicals are king. New Love is the release of hormones in the brain. Once these chemicals fade, what remains is true lov.e If you keep pumping someone full of these chemcials, the NEw Love won't fade. Therefore, you can make someone fall into love.

In order to make them fall out, if you can find how to elimiante these chemicals, you would force someone to fall out of love, and possible lose all happiness